Merge branch 'AUTOMATIC1111:master' into master

.github/workflows/run_tests.yaml

@@ -0,0 +1,31 @@
+name: Run basic features tests on CPU with empty SD model
+
+on:
+  - push
+  - pull_request
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v3
+      - name: Set up Python 3.10
+        uses: actions/setup-python@v4
+        with:
+          python-version: 3.10.6
+      - uses: actions/cache@v3
+        with:
+          path: ~/.cache/pip
+          key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
+          restore-keys: ${{ runner.os }}-pip-
+      - name: Run tests
+        run: python launch.py --tests basic_features --no-half --disable-opt-split-attention --use-cpu all --skip-torch-cuda-test
+      - name: Upload main app stdout-stderr
+        uses: actions/upload-artifact@v3
+        if: always()
+        with:
+          name: stdout-stderr
+          path: |
+            test/stdout.txt
+            test/stderr.txt

.gitignore

@@ -1,5 +1,6 @@
 __pycache__
 *.ckpt
+*.safetensors
 *.pth
 /ESRGAN/*
 /SwinIR/*

README.md

@@ -70,7 +70,7 @@ Check the [custom scripts](https://github.com/AUTOMATIC1111/stable-diffusion-web
      - separate prompts using uppercase `AND`
      - also supports weights for prompts: `a cat :1.2 AND a dog AND a penguin :2.2`
 - No token limit for prompts (original stable diffusion lets you use up to 75 tokens)
-- DeepDanbooru integration, creates danbooru style tags for anime prompts (add --deepdanbooru to commandline args)
+- DeepDanbooru integration, creates danbooru style tags for anime prompts
 - [xformers](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Xformers), major speed increase for select cards: (add --xformers to commandline args)
 - via extension: [History tab](https://github.com/yfszzx/stable-diffusion-webui-images-browser): view, direct and delete images conveniently within the UI
 - Generate forever option
@@ -84,26 +84,7 @@ Check the [custom scripts](https://github.com/AUTOMATIC1111/stable-diffusion-web
 - API
 - Support for dedicated [inpainting model](https://github.com/runwayml/stable-diffusion#inpainting-with-stable-diffusion) by RunwayML. 
 - via extension: [Aesthetic Gradients](https://github.com/AUTOMATIC1111/stable-diffusion-webui-aesthetic-gradients), a way to generate images with a specific aesthetic by using clip images embds (implementation of [https://github.com/vicgalle/stable-diffusion-aesthetic-gradients](https://github.com/vicgalle/stable-diffusion-aesthetic-gradients))
-
-## Where are Aesthetic Gradients?!?!
-Aesthetic Gradients are now an extension. You can install it using git:
-
-```commandline
-git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui-aesthetic-gradients extensions/aesthetic-gradients
-```
-
-After running this command, make sure that you have `aesthetic-gradients` dir in webui's `extensions` directory and restart
-the UI. The interface for Aesthetic Gradients should appear exactly the same as it was.
-
-## Where is History/Image browser?!?!
-Image browser is now an extension. You can install it using git:
-
-```commandline
-git clone https://github.com/yfszzx/stable-diffusion-webui-images-browser extensions/images-browser
-```
-
-After running this command, make sure that you have `images-browser` dir in webui's `extensions` directory and restart
-the UI. The interface for Image browser should appear exactly the same as it was.
+- [Stable Diffusion 2.0](https://github.com/Stability-AI/stablediffusion) support - see [wiki](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#stable-diffusion-20) for instructions
 
 ## Installation and Running
 Make sure the required [dependencies](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Dependencies) are met and follow the instructions available for both [NVidia](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Install-and-Run-on-NVidia-GPUs) (recommended) and [AMD](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Install-and-Run-on-AMD-GPUs) GPUs.
@@ -154,6 +135,7 @@ The documentation was moved from this README over to the project's [wiki](https:
 - SwinIR - https://github.com/JingyunLiang/SwinIR
 - Swin2SR - https://github.com/mv-lab/swin2sr
 - LDSR - https://github.com/Hafiidz/latent-diffusion
+- MiDaS - https://github.com/isl-org/MiDaS
 - Ideas for optimizations - https://github.com/basujindal/stable-diffusion
 - Cross Attention layer optimization - Doggettx - https://github.com/Doggettx/stable-diffusion, original idea for prompt editing.
 - Cross Attention layer optimization - InvokeAI, lstein - https://github.com/invoke-ai/InvokeAI (originally http://github.com/lstein/stable-diffusion)

extensions-builtin/LDSR/ldsr_model_arch.py

@@ -22,6 +22,7 @@ class LDSR:
         pl_sd = torch.load(self.modelPath, map_location="cpu")
         sd = pl_sd["state_dict"]
         config = OmegaConf.load(self.yamlPath)
+        config.model.target = "ldm.models.diffusion.ddpm.LatentDiffusionV1"
         model = instantiate_from_config(config.model)
         model.load_state_dict(sd, strict=False)
         model.cuda()

extensions-builtin/LDSR/preload.py

@@ -0,0 +1,6 @@
+import os
+from modules import paths
+
+
+def preload(parser):
+    parser.add_argument("--ldsr-models-path", type=str, help="Path to directory with LDSR model file(s).", default=os.path.join(paths.models_path, 'LDSR'))

extensions-builtin/LDSR/scripts/ldsr_model.py

@@ -5,8 +5,9 @@ import traceback
 from basicsr.utils.download_util import load_file_from_url
 
 from modules.upscaler import Upscaler, UpscalerData
-from modules.ldsr_model_arch import LDSR
-from modules import shared
+from ldsr_model_arch import LDSR
+from modules import shared, script_callbacks
+import sd_hijack_autoencoder, sd_hijack_ddpm_v1
 
 
 class UpscalerLDSR(Upscaler):
@@ -52,3 +53,12 @@ class UpscalerLDSR(Upscaler):
             return img
         ddim_steps = shared.opts.ldsr_steps
         return ldsr.super_resolution(img, ddim_steps, self.scale)
+
+
+def on_ui_settings():
+    import gradio as gr
+
+    shared.opts.add_option("ldsr_steps", shared.OptionInfo(100, "LDSR processing steps. Lower = faster", gr.Slider, {"minimum": 1, "maximum": 200, "step": 1}, section=('upscaling', "Upscaling")))
+
+
+script_callbacks.on_ui_settings(on_ui_settings)

extensions-builtin/LDSR/sd_hijack_autoencoder.py

@@ -0,0 +1,286 @@
+# The content of this file comes from the ldm/models/autoencoder.py file of the compvis/stable-diffusion repo
+# The VQModel & VQModelInterface were subsequently removed from ldm/models/autoencoder.py when we moved to the stability-ai/stablediffusion repo
+# As the LDSR upscaler relies on VQModel & VQModelInterface, the hijack aims to put them back into the ldm.models.autoencoder
+
+import torch
+import pytorch_lightning as pl
+import torch.nn.functional as F
+from contextlib import contextmanager
+from taming.modules.vqvae.quantize import VectorQuantizer2 as VectorQuantizer
+from ldm.modules.diffusionmodules.model import Encoder, Decoder
+from ldm.util import instantiate_from_config
+
+import ldm.models.autoencoder
+
+class VQModel(pl.LightningModule):
+    def __init__(self,
+                 ddconfig,
+                 lossconfig,
+                 n_embed,
+                 embed_dim,
+                 ckpt_path=None,
+                 ignore_keys=[],
+                 image_key="image",
+                 colorize_nlabels=None,
+                 monitor=None,
+                 batch_resize_range=None,
+                 scheduler_config=None,
+                 lr_g_factor=1.0,
+                 remap=None,
+                 sane_index_shape=False, # tell vector quantizer to return indices as bhw
+                 use_ema=False
+                 ):
+        super().__init__()
+        self.embed_dim = embed_dim
+        self.n_embed = n_embed
+        self.image_key = image_key
+        self.encoder = Encoder(**ddconfig)
+        self.decoder = Decoder(**ddconfig)
+        self.loss = instantiate_from_config(lossconfig)
+        self.quantize = VectorQuantizer(n_embed, embed_dim, beta=0.25,
+                                        remap=remap,
+                                        sane_index_shape=sane_index_shape)
+        self.quant_conv = torch.nn.Conv2d(ddconfig["z_channels"], embed_dim, 1)
+        self.post_quant_conv = torch.nn.Conv2d(embed_dim, ddconfig["z_channels"], 1)
+        if colorize_nlabels is not None:
+            assert type(colorize_nlabels)==int
+            self.register_buffer("colorize", torch.randn(3, colorize_nlabels, 1, 1))
+        if monitor is not None:
+            self.monitor = monitor
+        self.batch_resize_range = batch_resize_range
+        if self.batch_resize_range is not None:
+            print(f"{self.__class__.__name__}: Using per-batch resizing in range {batch_resize_range}.")
+
+        self.use_ema = use_ema
+        if self.use_ema:
+            self.model_ema = LitEma(self)
+            print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.")
+
+        if ckpt_path is not None:
+            self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys)
+        self.scheduler_config = scheduler_config
+        self.lr_g_factor = lr_g_factor
+
+    @contextmanager
+    def ema_scope(self, context=None):
+        if self.use_ema:
+            self.model_ema.store(self.parameters())
+            self.model_ema.copy_to(self)
+            if context is not None:
+                print(f"{context}: Switched to EMA weights")
+        try:
+            yield None
+        finally:
+            if self.use_ema:
+                self.model_ema.restore(self.parameters())
+                if context is not None:
+                    print(f"{context}: Restored training weights")
+
+    def init_from_ckpt(self, path, ignore_keys=list()):
+        sd = torch.load(path, map_location="cpu")["state_dict"]
+        keys = list(sd.keys())
+        for k in keys:
+            for ik in ignore_keys:
+                if k.startswith(ik):
+                    print("Deleting key {} from state_dict.".format(k))
+                    del sd[k]
+        missing, unexpected = self.load_state_dict(sd, strict=False)
+        print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys")
+        if len(missing) > 0:
+            print(f"Missing Keys: {missing}")
+            print(f"Unexpected Keys: {unexpected}")
+
+    def on_train_batch_end(self, *args, **kwargs):
+        if self.use_ema:
+            self.model_ema(self)
+
+    def encode(self, x):
+        h = self.encoder(x)
+        h = self.quant_conv(h)
+        quant, emb_loss, info = self.quantize(h)
+        return quant, emb_loss, info
+
+    def encode_to_prequant(self, x):
+        h = self.encoder(x)
+        h = self.quant_conv(h)
+        return h
+
+    def decode(self, quant):
+        quant = self.post_quant_conv(quant)
+        dec = self.decoder(quant)
+        return dec
+
+    def decode_code(self, code_b):
+        quant_b = self.quantize.embed_code(code_b)
+        dec = self.decode(quant_b)
+        return dec
+
+    def forward(self, input, return_pred_indices=False):
+        quant, diff, (_,_,ind) = self.encode(input)
+        dec = self.decode(quant)
+        if return_pred_indices:
+            return dec, diff, ind
+        return dec, diff
+
+    def get_input(self, batch, k):
+        x = batch[k]
+        if len(x.shape) == 3:
+            x = x[..., None]
+        x = x.permute(0, 3, 1, 2).to(memory_format=torch.contiguous_format).float()
+        if self.batch_resize_range is not None:
+            lower_size = self.batch_resize_range[0]
+            upper_size = self.batch_resize_range[1]
+            if self.global_step <= 4:
+                # do the first few batches with max size to avoid later oom
+                new_resize = upper_size
+            else:
+                new_resize = np.random.choice(np.arange(lower_size, upper_size+16, 16))
+            if new_resize != x.shape[2]:
+                x = F.interpolate(x, size=new_resize, mode="bicubic")
+            x = x.detach()
+        return x
+
+    def training_step(self, batch, batch_idx, optimizer_idx):
+        # https://github.com/pytorch/pytorch/issues/37142
+        # try not to fool the heuristics
+        x = self.get_input(batch, self.image_key)
+        xrec, qloss, ind = self(x, return_pred_indices=True)
+
+        if optimizer_idx == 0:
+            # autoencode
+            aeloss, log_dict_ae = self.loss(qloss, x, xrec, optimizer_idx, self.global_step,
+                                            last_layer=self.get_last_layer(), split="train",
+                                            predicted_indices=ind)
+
+            self.log_dict(log_dict_ae, prog_bar=False, logger=True, on_step=True, on_epoch=True)
+            return aeloss
+
+        if optimizer_idx == 1:
+            # discriminator
+            discloss, log_dict_disc = self.loss(qloss, x, xrec, optimizer_idx, self.global_step,
+                                            last_layer=self.get_last_layer(), split="train")
+            self.log_dict(log_dict_disc, prog_bar=False, logger=True, on_step=True, on_epoch=True)
+            return discloss
+
+    def validation_step(self, batch, batch_idx):
+        log_dict = self._validation_step(batch, batch_idx)
+        with self.ema_scope():
+            log_dict_ema = self._validation_step(batch, batch_idx, suffix="_ema")
+        return log_dict
+
+    def _validation_step(self, batch, batch_idx, suffix=""):
+        x = self.get_input(batch, self.image_key)
+        xrec, qloss, ind = self(x, return_pred_indices=True)
+        aeloss, log_dict_ae = self.loss(qloss, x, xrec, 0,
+                                        self.global_step,
+                                        last_layer=self.get_last_layer(),
+                                        split="val"+suffix,
+                                        predicted_indices=ind
+                                        )
+
+        discloss, log_dict_disc = self.loss(qloss, x, xrec, 1,
+                                            self.global_step,
+                                            last_layer=self.get_last_layer(),
+                                            split="val"+suffix,
+                                            predicted_indices=ind
+                                            )
+        rec_loss = log_dict_ae[f"val{suffix}/rec_loss"]
+        self.log(f"val{suffix}/rec_loss", rec_loss,
+                   prog_bar=True, logger=True, on_step=False, on_epoch=True, sync_dist=True)
+        self.log(f"val{suffix}/aeloss", aeloss,
+                   prog_bar=True, logger=True, on_step=False, on_epoch=True, sync_dist=True)
+        if version.parse(pl.__version__) >= version.parse('1.4.0'):
+            del log_dict_ae[f"val{suffix}/rec_loss"]
+        self.log_dict(log_dict_ae)
+        self.log_dict(log_dict_disc)
+        return self.log_dict
+
+    def configure_optimizers(self):
+        lr_d = self.learning_rate
+        lr_g = self.lr_g_factor*self.learning_rate
+        print("lr_d", lr_d)
+        print("lr_g", lr_g)
+        opt_ae = torch.optim.Adam(list(self.encoder.parameters())+
+                                  list(self.decoder.parameters())+
+                                  list(self.quantize.parameters())+
+                                  list(self.quant_conv.parameters())+
+                                  list(self.post_quant_conv.parameters()),
+                                  lr=lr_g, betas=(0.5, 0.9))
+        opt_disc = torch.optim.Adam(self.loss.discriminator.parameters(),
+                                    lr=lr_d, betas=(0.5, 0.9))
+
+        if self.scheduler_config is not None:
+            scheduler = instantiate_from_config(self.scheduler_config)
+
+            print("Setting up LambdaLR scheduler...")
+            scheduler = [
+                {
+                    'scheduler': LambdaLR(opt_ae, lr_lambda=scheduler.schedule),
+                    'interval': 'step',
+                    'frequency': 1
+                },
+                {
+                    'scheduler': LambdaLR(opt_disc, lr_lambda=scheduler.schedule),
+                    'interval': 'step',
+                    'frequency': 1
+                },
+            ]
+            return [opt_ae, opt_disc], scheduler
+        return [opt_ae, opt_disc], []
+
+    def get_last_layer(self):
+        return self.decoder.conv_out.weight
+
+    def log_images(self, batch, only_inputs=False, plot_ema=False, **kwargs):
+        log = dict()
+        x = self.get_input(batch, self.image_key)
+        x = x.to(self.device)
+        if only_inputs:
+            log["inputs"] = x
+            return log
+        xrec, _ = self(x)
+        if x.shape[1] > 3:
+            # colorize with random projection
+            assert xrec.shape[1] > 3
+            x = self.to_rgb(x)
+            xrec = self.to_rgb(xrec)
+        log["inputs"] = x
+        log["reconstructions"] = xrec
+        if plot_ema:
+            with self.ema_scope():
+                xrec_ema, _ = self(x)
+                if x.shape[1] > 3: xrec_ema = self.to_rgb(xrec_ema)
+                log["reconstructions_ema"] = xrec_ema
+        return log
+
+    def to_rgb(self, x):
+        assert self.image_key == "segmentation"
+        if not hasattr(self, "colorize"):
+            self.register_buffer("colorize", torch.randn(3, x.shape[1], 1, 1).to(x))
+        x = F.conv2d(x, weight=self.colorize)
+        x = 2.*(x-x.min())/(x.max()-x.min()) - 1.
+        return x
+
+
+class VQModelInterface(VQModel):
+    def __init__(self, embed_dim, *args, **kwargs):
+        super().__init__(embed_dim=embed_dim, *args, **kwargs)
+        self.embed_dim = embed_dim
+
+    def encode(self, x):
+        h = self.encoder(x)
+        h = self.quant_conv(h)
+        return h
+
+    def decode(self, h, force_not_quantize=False):
+        # also go through quantization layer
+        if not force_not_quantize:
+            quant, emb_loss, info = self.quantize(h)
+        else:
+            quant = h
+        quant = self.post_quant_conv(quant)
+        dec = self.decoder(quant)
+        return dec
+
+setattr(ldm.models.autoencoder, "VQModel", VQModel)
+setattr(ldm.models.autoencoder, "VQModelInterface", VQModelInterface)

extensions-builtin/ScuNET/preload.py

@@ -0,0 +1,6 @@
+import os
+from modules import paths
+
+
+def preload(parser):
+    parser.add_argument("--scunet-models-path", type=str, help="Path to directory with ScuNET model file(s).", default=os.path.join(paths.models_path, 'ScuNET'))

extensions-builtin/ScuNET/scripts/scunet_model.py

@@ -9,7 +9,7 @@ from basicsr.utils.download_util import load_file_from_url
 
 import modules.upscaler
 from modules import devices, modelloader
-from modules.scunet_model_arch import SCUNet as net
+from scunet_model_arch import SCUNet as net
 
 
 class UpscalerScuNET(modules.upscaler.Upscaler):
@@ -49,12 +49,12 @@ class UpscalerScuNET(modules.upscaler.Upscaler):
         if model is None:
             return img
 
-        device = devices.device_scunet
+        device = devices.get_device_for('scunet')
         img = np.array(img)
         img = img[:, :, ::-1]
         img = np.moveaxis(img, 2, 0) / 255
         img = torch.from_numpy(img).float()
-        img = devices.mps_contiguous_to(img.unsqueeze(0), device)
+        img = img.unsqueeze(0).to(device)
 
         with torch.no_grad():
             output = model(img)
@@ -66,7 +66,7 @@ class UpscalerScuNET(modules.upscaler.Upscaler):
         return PIL.Image.fromarray(output, 'RGB')
 
     def load_model(self, path: str):
-        device = devices.device_scunet
+        device = devices.get_device_for('scunet')
         if "http" in path:
             filename = load_file_from_url(url=self.model_url, model_dir=self.model_path, file_name="%s.pth" % self.name,
                                           progress=True)

extensions-builtin/SwinIR/preload.py

@@ -0,0 +1,6 @@
+import os
+from modules import paths
+
+
+def preload(parser):
+    parser.add_argument("--swinir-models-path", type=str, help="Path to directory with SwinIR model file(s).", default=os.path.join(paths.models_path, 'SwinIR'))

extensions-builtin/SwinIR/scripts/swinir_model.py

@@ -7,15 +7,14 @@ from PIL import Image
 from basicsr.utils.download_util import load_file_from_url
 from tqdm import tqdm
 
-from modules import modelloader, devices
+from modules import modelloader, devices, script_callbacks, shared
 from modules.shared import cmd_opts, opts
-from modules.swinir_model_arch import SwinIR as net
-from modules.swinir_model_arch_v2 import Swin2SR as net2
+from swinir_model_arch import SwinIR as net
+from swinir_model_arch_v2 import Swin2SR as net2
 from modules.upscaler import Upscaler, UpscalerData
 
-precision_scope = (
-    torch.autocast if cmd_opts.precision == "autocast" else contextlib.nullcontext
-)
+
+device_swinir = devices.get_device_for('swinir')
 
 
 class UpscalerSwinIR(Upscaler):
@@ -42,7 +41,7 @@ class UpscalerSwinIR(Upscaler):
         model = self.load_model(model_file)
         if model is None:
             return img
-        model = model.to(devices.device_swinir)
+        model = model.to(device_swinir, dtype=devices.dtype)
         img = upscale(img, model)
         try:
             torch.cuda.empty_cache()
@@ -94,25 +93,27 @@ class UpscalerSwinIR(Upscaler):
             model.load_state_dict(pretrained_model[params], strict=True)
         else:
             model.load_state_dict(pretrained_model, strict=True)
-        if not cmd_opts.no_half:
-            model = model.half()
         return model
 
 
 def upscale(
         img,
         model,
-        tile=opts.SWIN_tile,
-        tile_overlap=opts.SWIN_tile_overlap,
+        tile=None,
+        tile_overlap=None,
         window_size=8,
         scale=4,
 ):
+    tile = tile or opts.SWIN_tile
+    tile_overlap = tile_overlap or opts.SWIN_tile_overlap
+
+
     img = np.array(img)
     img = img[:, :, ::-1]
     img = np.moveaxis(img, 2, 0) / 255
     img = torch.from_numpy(img).float()
-    img = devices.mps_contiguous_to(img.unsqueeze(0), devices.device_swinir)
-    with torch.no_grad(), precision_scope("cuda"):
+    img = img.unsqueeze(0).to(device_swinir, dtype=devices.dtype)
+    with torch.no_grad(), devices.autocast():
         _, _, h_old, w_old = img.size()
         h_pad = (h_old // window_size + 1) * window_size - h_old
         w_pad = (w_old // window_size + 1) * window_size - w_old
@@ -139,8 +140,8 @@ def inference(img, model, tile, tile_overlap, window_size, scale):
     stride = tile - tile_overlap
     h_idx_list = list(range(0, h - tile, stride)) + [h - tile]
     w_idx_list = list(range(0, w - tile, stride)) + [w - tile]
-    E = torch.zeros(b, c, h * sf, w * sf, dtype=torch.half, device=devices.device_swinir).type_as(img)
-    W = torch.zeros_like(E, dtype=torch.half, device=devices.device_swinir)
+    E = torch.zeros(b, c, h * sf, w * sf, dtype=devices.dtype, device=device_swinir).type_as(img)
+    W = torch.zeros_like(E, dtype=devices.dtype, device=device_swinir)
 
     with tqdm(total=len(h_idx_list) * len(w_idx_list), desc="SwinIR tiles") as pbar:
         for h_idx in h_idx_list:
@@ -159,3 +160,13 @@ def inference(img, model, tile, tile_overlap, window_size, scale):
     output = E.div_(W)
 
     return output
+
+
+def on_ui_settings():
+    import gradio as gr
+
+    shared.opts.add_option("SWIN_tile", shared.OptionInfo(192, "Tile size for all SwinIR.", gr.Slider, {"minimum": 16, "maximum": 512, "step": 16}, section=('upscaling', "Upscaling")))
+    shared.opts.add_option("SWIN_tile_overlap", shared.OptionInfo(8, "Tile overlap, in pixels for SwinIR. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}, section=('upscaling', "Upscaling")))
+
+
+script_callbacks.on_ui_settings(on_ui_settings)

extensions-builtin/prompt-bracket-checker/javascript/prompt-bracket-checker.js

@@ -0,0 +1,107 @@
+// Stable Diffusion WebUI - Bracket checker
+// Version 1.0
+// By Hingashi no Florin/Bwin4L
+// Counts open and closed brackets (round, square, curly) in the prompt and negative prompt text boxes in the txt2img and img2img tabs.
+// If there's a mismatch, the keyword counter turns red and if you hover on it, a tooltip tells you what's wrong.
+
+function checkBrackets(evt) {
+  textArea = evt.target;
+  tabName = evt.target.parentElement.parentElement.id.split("_")[0];
+  counterElt = document.querySelector('gradio-app').shadowRoot.querySelector('#' + tabName + '_token_counter');
+
+  promptName = evt.target.parentElement.parentElement.id.includes('neg') ? ' negative' : '';
+
+  errorStringParen  = '(' + tabName + promptName + ' prompt) - Different number of opening and closing parentheses detected.\n';
+  errorStringSquare = '[' + tabName + promptName + ' prompt] - Different number of opening and closing square brackets detected.\n';
+  errorStringCurly  = '{' + tabName + promptName + ' prompt} - Different number of opening and closing curly brackets detected.\n';
+
+  openBracketRegExp = /\(/g;
+  closeBracketRegExp = /\)/g;
+
+  openSquareBracketRegExp = /\[/g;
+  closeSquareBracketRegExp = /\]/g;
+
+  openCurlyBracketRegExp = /\{/g;
+  closeCurlyBracketRegExp = /\}/g;
+
+  totalOpenBracketMatches = 0;
+  totalCloseBracketMatches = 0;
+  totalOpenSquareBracketMatches = 0;
+  totalCloseSquareBracketMatches = 0;
+  totalOpenCurlyBracketMatches = 0;
+  totalCloseCurlyBracketMatches = 0;
+
+  openBracketMatches = textArea.value.match(openBracketRegExp);
+  if(openBracketMatches) {
+    totalOpenBracketMatches = openBracketMatches.length;
+  }
+
+  closeBracketMatches = textArea.value.match(closeBracketRegExp);
+  if(closeBracketMatches) {
+    totalCloseBracketMatches = closeBracketMatches.length;
+  }
+
+  openSquareBracketMatches = textArea.value.match(openSquareBracketRegExp);
+  if(openSquareBracketMatches) {
+    totalOpenSquareBracketMatches = openSquareBracketMatches.length;
+  }
+
+  closeSquareBracketMatches = textArea.value.match(closeSquareBracketRegExp);
+  if(closeSquareBracketMatches) {
+    totalCloseSquareBracketMatches = closeSquareBracketMatches.length;
+  }
+
+  openCurlyBracketMatches = textArea.value.match(openCurlyBracketRegExp);
+  if(openCurlyBracketMatches) {
+    totalOpenCurlyBracketMatches = openCurlyBracketMatches.length;
+  }
+
+  closeCurlyBracketMatches = textArea.value.match(closeCurlyBracketRegExp);
+  if(closeCurlyBracketMatches) {
+    totalCloseCurlyBracketMatches = closeCurlyBracketMatches.length;
+  }
+
+  if(totalOpenBracketMatches != totalCloseBracketMatches) {
+    if(!counterElt.title.includes(errorStringParen)) {
+      counterElt.title += errorStringParen;
+    }
+  } else {
+    counterElt.title = counterElt.title.replace(errorStringParen, '');
+  }
+
+  if(totalOpenSquareBracketMatches != totalCloseSquareBracketMatches) {
+    if(!counterElt.title.includes(errorStringSquare)) {
+      counterElt.title += errorStringSquare;
+    }
+  } else {
+    counterElt.title = counterElt.title.replace(errorStringSquare, '');
+  }
+
+  if(totalOpenCurlyBracketMatches != totalCloseCurlyBracketMatches) {
+    if(!counterElt.title.includes(errorStringCurly)) {
+      counterElt.title += errorStringCurly;
+    }
+  } else {
+    counterElt.title = counterElt.title.replace(errorStringCurly, '');
+  }
+
+  if(counterElt.title != '') {
+    counterElt.style = 'color: #FF5555;';
+  } else {
+    counterElt.style = 'color: #000;';
+  }
+}
+
+var shadowRootLoaded = setInterval(function() {
+  var shadowTextArea = document.querySelector('gradio-app').shadowRoot.querySelectorAll('#txt2img_prompt > label > textarea');
+  if(shadowTextArea.length < 1) {
+      return false;
+  }
+
+  clearInterval(shadowRootLoaded);
+
+  document.querySelector('gradio-app').shadowRoot.querySelector('#txt2img_prompt').onkeyup = checkBrackets;
+  document.querySelector('gradio-app').shadowRoot.querySelector('#txt2img_neg_prompt').onkeyup = checkBrackets;
+  document.querySelector('gradio-app').shadowRoot.querySelector('#img2img_prompt').onkeyup = checkBrackets;
+  document.querySelector('gradio-app').shadowRoot.querySelector('#img2img_neg_prompt').onkeyup = checkBrackets;
+}, 1000);

javascript/generationParams.js

@@ -0,0 +1,33 @@
+// attaches listeners to the txt2img and img2img galleries to update displayed generation param text when the image changes
+
+let txt2img_gallery, img2img_gallery, modal = undefined;
+onUiUpdate(function(){
+	if (!txt2img_gallery) {
+		txt2img_gallery = attachGalleryListeners("txt2img")
+	}
+	if (!img2img_gallery) {
+		img2img_gallery = attachGalleryListeners("img2img")
+	}
+	if (!modal) {
+		modal = gradioApp().getElementById('lightboxModal')
+		modalObserver.observe(modal,  { attributes : true, attributeFilter : ['style'] });
+	}
+});
+
+let modalObserver = new MutationObserver(function(mutations) {
+	mutations.forEach(function(mutationRecord) {
+		let selectedTab = gradioApp().querySelector('#tabs div button.bg-white')?.innerText
+		if (mutationRecord.target.style.display === 'none' && selectedTab === 'txt2img' || selectedTab === 'img2img')
+			gradioApp().getElementById(selectedTab+"_generation_info_button").click()
+	});
+});
+
+function attachGalleryListeners(tab_name) {
+	gallery = gradioApp().querySelector('#'+tab_name+'_gallery')
+	gallery?.addEventListener('click', () => gradioApp().getElementById(tab_name+"_generation_info_button").click());
+	gallery?.addEventListener('keydown', (e) => {
+		if (e.keyCode == 37 || e.keyCode == 39) // left or right arrow
+			gradioApp().getElementById(tab_name+"_generation_info_button").click()
+	});
+	return gallery;
+}

javascript/hints.js

@@ -63,8 +63,8 @@ titles = {
 
     "Interrogate": "Reconstruct prompt from existing image and put it into the prompt field.",
 
-    "Images filename pattern": "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.",
-    "Directory name pattern": "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.",
+    "Images filename pattern": "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [model_name], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.",
+    "Directory name pattern": "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [model_name], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.",
     "Max prompt words": "Set the maximum number of words to be used in the [prompt_words] option; ATTENTION: If the words are too long, they may exceed the maximum length of the file path that the system can handle",
 
     "Loopback": "Process an image, use it as an input, repeat.",
@@ -95,6 +95,8 @@ titles = {
     "Add difference": "Result = A + (B - C) * M",
 
     "Learning rate": "how fast should the training go. Low values will take longer to train, high values may fail to converge (not generate accurate results) and/or may break the embedding (This has happened if you see Loss: nan in the training info textbox. If this happens, you need to manually restore your embedding from an older not-broken backup).\n\nYou can set a single numeric value, or multiple learning rates using the syntax:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nWill train with rate of 0.005 for first 100 steps, then 1e-3 until 1000 steps, then 1e-5 for all remaining steps.",
+
+    "Clip skip": "Early stopping parameter for CLIP model; 1 is stop at last layer as usual, 2 is stop at penultimate layer, etc."
 }
 
 

javascript/progressbar.js

@@ -23,7 +23,7 @@ function check_progressbar(id_part, id_progressbar, id_progressbar_span, id_skip
     
     if(opts.show_progress_in_title && progressbar && progressbar.offsetParent){
         if(progressbar.innerText){
-            let newtitle = 'Stable Diffusion - ' + progressbar.innerText
+            let newtitle = '[' + progressbar.innerText.trim() + '] Stable Diffusion';
             if(document.title != newtitle){
                 document.title =  newtitle;          
             }
@@ -92,14 +92,26 @@ function check_gallery(id_gallery){
             if (prevSelectedIndex !== -1 && galleryButtons.length>prevSelectedIndex && !galleryBtnSelected) {
                 // automatically re-open previously selected index (if exists)
                 activeElement = gradioApp().activeElement;
+                let scrollX = window.scrollX;
+                let scrollY = window.scrollY;
 
                 galleryButtons[prevSelectedIndex].click();
                 showGalleryImage();
 
+                // When the gallery button is clicked, it gains focus and scrolls itself into view
+                // We need to scroll back to the previous position
+                setTimeout(function (){
+                    window.scrollTo(scrollX, scrollY);
+                }, 50);
+
                 if(activeElement){
                     // i fought this for about an hour; i don't know why the focus is lost or why this helps recover it
-                    // if somenoe has a better solution please by all means
-                    setTimeout(function() { activeElement.focus() }, 1);
+                    // if someone has a better solution please by all means
+                    setTimeout(function (){
+                        activeElement.focus({
+                            preventScroll: true // Refocus the element that was focused before the gallery was opened without scrolling to it
+                        })
+                    }, 1);
                 }
             }
         })

javascript/ui.js

@@ -8,8 +8,8 @@ function set_theme(theme){
 }
 
 function selected_gallery_index(){
-    var buttons = gradioApp().querySelectorAll('[style="display: block;"].tabitem .gallery-item')
-    var button = gradioApp().querySelector('[style="display: block;"].tabitem .gallery-item.\\!ring-2')
+    var buttons = gradioApp().querySelectorAll('[style="display: block;"].tabitem div[id$=_gallery] .gallery-item')
+    var button = gradioApp().querySelector('[style="display: block;"].tabitem div[id$=_gallery] .gallery-item.\\!ring-2')
 
     var result = -1
     buttons.forEach(function(v, i){ if(v==button) { result = i } })

launch.py

@@ -5,6 +5,8 @@ import sys
 import importlib.util
 import shlex
 import platform
+import argparse
+import json
 
 dir_repos = "repositories"
 dir_extensions = "extensions"
@@ -17,6 +19,19 @@ def extract_arg(args, name):
     return [x for x in args if x != name], name in args
 
 
+def extract_opt(args, name):
+    opt = None
+    is_present = False
+    if name in args:
+        is_present = True
+        idx = args.index(name)
+        del args[idx]
+        if idx < len(args) and args[idx][0] != "-":
+            opt = args[idx]
+            del args[idx]
+    return args, is_present, opt
+
+
 def run(command, desc=None, errdesc=None, custom_env=None):
     if desc is not None:
         print(desc)
@@ -105,22 +120,41 @@ def version_check(commit):
         print("version check failed", e)
 
 
-def run_extensions_installers():
-    if not os.path.isdir(dir_extensions):
+def run_extension_installer(extension_dir):
+    path_installer = os.path.join(extension_dir, "install.py")
+    if not os.path.isfile(path_installer):
         return
 
-    for dirname_extension in os.listdir(dir_extensions):
-        path_installer = os.path.join(dir_extensions, dirname_extension, "install.py")
-        if not os.path.isfile(path_installer):
-            continue
+    try:
+        env = os.environ.copy()
+        env['PYTHONPATH'] = os.path.abspath(".")
+
+        print(run(f'"{python}" "{path_installer}"', errdesc=f"Error running install.py for extension {extension_dir}", custom_env=env))
+    except Exception as e:
+        print(e, file=sys.stderr)
+
+
+def list_extensions(settings_file):
+    settings = {}
+
+    try:
+        if os.path.isfile(settings_file):
+            with open(settings_file, "r", encoding="utf8") as file:
+                settings = json.load(file)
+    except Exception as e:
+        print(e, file=sys.stderr)
+
+    disabled_extensions = set(settings.get('disabled_extensions', []))
+
+    return [x for x in os.listdir(dir_extensions) if x not in disabled_extensions]
 
-        try:
-            env = os.environ.copy()
-            env['PYTHONPATH'] = os.path.abspath(".")
 
-            print(run(f'"{python}" "{path_installer}"', errdesc=f"Error running install.py for extension {dirname_extension}", custom_env=env))
-        except Exception as e:
-            print(e, file=sys.stderr)
+def run_extensions_installers(settings_file):
+    if not os.path.isdir(dir_extensions):
+        return
+
+    for dirname_extension in list_extensions(settings_file):
+        run_extension_installer(os.path.join(dir_extensions, dirname_extension))
 
 
 def prepare_enviroment():
@@ -130,31 +164,34 @@ def prepare_enviroment():
 
     gfpgan_package = os.environ.get('GFPGAN_PACKAGE', "git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379")
     clip_package = os.environ.get('CLIP_PACKAGE', "git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1")
-    deepdanbooru_package = os.environ.get('DEEPDANBOORU_PACKAGE', "git+https://github.com/KichangKim/DeepDanbooru.git@d91a2963bf87c6a770d74894667e9ffa9f6de7ff")
+    openclip_package = os.environ.get('OPENCLIP_PACKAGE', "git+https://github.com/mlfoundations/open_clip.git@bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b")
 
     xformers_windows_package = os.environ.get('XFORMERS_WINDOWS_PACKAGE', 'https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl')
 
-    stable_diffusion_repo = os.environ.get('STABLE_DIFFUSION_REPO', "https://github.com/CompVis/stable-diffusion.git")
-    taming_transformers_repo = os.environ.get('TAMING_REANSFORMERS_REPO', "https://github.com/CompVis/taming-transformers.git")
+    stable_diffusion_repo = os.environ.get('STABLE_DIFFUSION_REPO', "https://github.com/Stability-AI/stablediffusion.git")
+    taming_transformers_repo = os.environ.get('TAMING_TRANSFORMERS_REPO', "https://github.com/CompVis/taming-transformers.git")
     k_diffusion_repo = os.environ.get('K_DIFFUSION_REPO', 'https://github.com/crowsonkb/k-diffusion.git')
-    codeformer_repo = os.environ.get('CODEFORMET_REPO', 'https://github.com/sczhou/CodeFormer.git')
+    codeformer_repo = os.environ.get('CODEFORMER_REPO', 'https://github.com/sczhou/CodeFormer.git')
     blip_repo = os.environ.get('BLIP_REPO', 'https://github.com/salesforce/BLIP.git')
 
-    stable_diffusion_commit_hash = os.environ.get('STABLE_DIFFUSION_COMMIT_HASH', "69ae4b35e0a0f6ee1af8bb9a5d0016ccb27e36dc")
+    stable_diffusion_commit_hash = os.environ.get('STABLE_DIFFUSION_COMMIT_HASH', "47b6b607fdd31875c9279cd2f4f16b92e4ea958e")
     taming_transformers_commit_hash = os.environ.get('TAMING_TRANSFORMERS_COMMIT_HASH', "24268930bf1dce879235a7fddd0b2355b84d7ea6")
-    k_diffusion_commit_hash = os.environ.get('K_DIFFUSION_COMMIT_HASH', "60e5042ca0da89c14d1dd59d73883280f8fce991")
+    k_diffusion_commit_hash = os.environ.get('K_DIFFUSION_COMMIT_HASH', "5b3af030dd83e0297272d861c19477735d0317ec")
     codeformer_commit_hash = os.environ.get('CODEFORMER_COMMIT_HASH', "c5b4593074ba6214284d6acd5f1719b6c5d739af")
     blip_commit_hash = os.environ.get('BLIP_COMMIT_HASH', "48211a1594f1321b00f14c9f7a5b4813144b2fb9")
 
     sys.argv += shlex.split(commandline_args)
-    test_argv = [x for x in sys.argv if x != '--tests']
 
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--ui-settings-file", type=str, help="filename to use for ui settings", default='config.json')
+    args, _ = parser.parse_known_args(sys.argv)
+
+    sys.argv, _ = extract_arg(sys.argv, '-f')
     sys.argv, skip_torch_cuda_test = extract_arg(sys.argv, '--skip-torch-cuda-test')
     sys.argv, reinstall_xformers = extract_arg(sys.argv, '--reinstall-xformers')
     sys.argv, update_check = extract_arg(sys.argv, '--update-check')
-    sys.argv, run_tests = extract_arg(sys.argv, '--tests')
+    sys.argv, run_tests, test_dir = extract_opt(sys.argv, '--tests')
     xformers = '--xformers' in sys.argv
-    deepdanbooru = '--deepdanbooru' in sys.argv
     ngrok = '--ngrok' in sys.argv
 
     try:
@@ -177,6 +214,9 @@ def prepare_enviroment():
     if not is_installed("clip"):
         run_pip(f"install {clip_package}", "clip")
 
+    if not is_installed("open_clip"):
+        run_pip(f"install {openclip_package}", "open_clip")
+
     if (not is_installed("xformers") or reinstall_xformers) and xformers:
         if platform.system() == "Windows":
             if platform.python_version().startswith("3.10"):
@@ -189,15 +229,12 @@ def prepare_enviroment():
         elif platform.system() == "Linux":
             run_pip("install xformers", "xformers")
 
-    if not is_installed("deepdanbooru") and deepdanbooru:
-        run_pip(f"install {deepdanbooru_package}#egg=deepdanbooru[tensorflow] tensorflow==2.10.0 tensorflow-io==0.27.0", "deepdanbooru")
-
     if not is_installed("pyngrok") and ngrok:
         run_pip("install pyngrok", "ngrok")
 
     os.makedirs(dir_repos, exist_ok=True)
 
-    git_clone(stable_diffusion_repo, repo_dir('stable-diffusion'), "Stable Diffusion", stable_diffusion_commit_hash)
+    git_clone(stable_diffusion_repo, repo_dir('stable-diffusion-stability-ai'), "Stable Diffusion", stable_diffusion_commit_hash)
     git_clone(taming_transformers_repo, repo_dir('taming-transformers'), "Taming Transformers", taming_transformers_commit_hash)
     git_clone(k_diffusion_repo, repo_dir('k-diffusion'), "K-diffusion", k_diffusion_commit_hash)
     git_clone(codeformer_repo, repo_dir('CodeFormer'), "CodeFormer", codeformer_commit_hash)
@@ -208,7 +245,7 @@ def prepare_enviroment():
 
     run_pip(f"install -r {requirements_file}", "requirements for Web UI")
 
-    run_extensions_installers()
+    run_extensions_installers(settings_file=args.ui_settings_file)
 
     if update_check:
         version_check(commit)
@@ -218,24 +255,30 @@ def prepare_enviroment():
         exit(0)
 
     if run_tests:
-        tests(test_argv)
-        exit(0)
+        exitcode = tests(test_dir)
+        exit(exitcode)
 
 
-def tests(argv):
-    if "--api" not in argv:
-        argv.append("--api")
+def tests(test_dir):
+    if "--api" not in sys.argv:
+        sys.argv.append("--api")
+    if "--ckpt" not in sys.argv:
+        sys.argv.append("--ckpt")
+        sys.argv.append("./test/test_files/empty.pt")
+    if "--skip-torch-cuda-test" not in sys.argv:
+        sys.argv.append("--skip-torch-cuda-test")
 
-    print(f"Launching Web UI in another process for testing with arguments: {' '.join(argv[1:])}")
+    print(f"Launching Web UI in another process for testing with arguments: {' '.join(sys.argv[1:])}")
 
     with open('test/stdout.txt', "w", encoding="utf8") as stdout, open('test/stderr.txt', "w", encoding="utf8") as stderr:
-        proc = subprocess.Popen([sys.executable, *argv], stdout=stdout, stderr=stderr)
+        proc = subprocess.Popen([sys.executable, *sys.argv], stdout=stdout, stderr=stderr)
 
     import test.server_poll
-    test.server_poll.run_tests()
+    exitcode = test.server_poll.run_tests(proc, test_dir)
 
     print(f"Stopping Web UI process with id {proc.pid}")
     proc.kill()
+    return exitcode
 
 
 def start():

localizations/ar_AR.json

@@ -1,518 +0,0 @@
-{
-  "rtl": true,
-  "Loading...": "لحظة...",
-  "view": "اعرض ",
-  "api": "واجهة البرمجة",
-  "built with gradio": "مبني باستخدام gradio",
-  "Stable Diffusion checkpoint": "أوزان نموذج الإنتشار المسقر",
-  "txt2img": "نص إلى صورة",
-  "Prompt": "الطلب",
-  "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "الطلب (لبدء الإنتاج Ctrl+Enter أو Alt+Enter اضغط)",
-  "Negative prompt": "عكس الطلب",
-  "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "عكس الطلب (لبدء الإنتاج Ctrl+Enter أو Alt+Enter اضغط)",
-  "Add a random artist to the prompt.": "أضف فنان عشوائي للطلب",
-  "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "اقرأ عوامل الإنتاج من الطلب أو من الإنتاج السابق إذا كان الطلب فارغا",
-  "Save style": "احتفظ بالطلب وعكسه كإضافة",
-  "Apply selected styles to current prompt": "ألحق الإضافات المحددة إلى الطلب وعكسه",
-  "Generate": "أنتج",
-  "Skip": "تخطى",
-  "Stop processing current image and continue processing.": "لا تكمل خطوات هذة الحزمة وانتقل إلى الحزمة التالية",
-  "Interrupt": "توقف",
-  "Stop processing images and return any results accumulated so far.": "توقف عن الإنتاج واعرض ما تم إلى الآن",
-  "Style 1": "الإضافة 1",
-  "Style to apply; styles have components for both positive and negative prompts and apply to both": "الإضافات (styles) عبارة عن كلمات تتكرر كثيرا يتم إلحاقها بالطلب وعكسه عند الرغبة",
-  "Style 2": "الإضافة 2",
-  "Do not do anything special": "لا يغير شيئا",
-  "Sampling Steps": "عدد الخطوات",
-  "Sampling method": "أسلوب الخطو",
-  "Which algorithm to use to produce the image": "Sampler: اسم نظام تحديد طريقة تغيير المسافات بين الخطوات",
-  "Euler a": "Euler a",
-  "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral: طريقة مبدعة يمكن أن تنتج صور مختلفة على حسب عدد الخطوات، لا تتغير بعد 30-40 خطوة",
-  "Euler": "Euler",
-  "LMS": "LMS",
-  "Heun": "Heun",
-  "DPM2": "DPM2",
-  "DPM2 a": "DPM2 a",
-  "DPM fast": "DPM fast",
-  "DPM adaptive": "DPM adaptive",
-  "LMS Karras": "LMS Karras",
-  "DPM2 Karras": "DPM2 Karras",
-  "DPM2 a Karras": "DPM2 a Karras",
-  "DDIM": "DDIM",
-  "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Models: الأفضل في الإنتاج الجزئي",
-  "PLMS": "PLMS",
-  "Width": "العرض",
-  "Height": "الإرتفاع",
-  "Restore faces": "تحسين الوجوه",
-  "Tiling": "ترصيف",
-  "Produce an image that can be tiled.": "أنتج صور يمكن ترصيفها بجانب بعضها كالبلاط",
-  "Highres. fix": "إصلاح الدقة العالية",
-  "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "أنتج صورة بدقة منخفضة ثم قم برفع الدقة فيما بعد لمنع التشوهات التي تحصل عندما تكون الدقة المطلوبة كبيرة",
-  "Firstpass width": "العرض الأولي",
-  "Firstpass height": "الإرتفاع الأولي",
-  "Denoising strength": "المدى",
-  "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Denoising strength: حدد مدى الإبتعاد عن الصورة (عدد الخطوات الفعلي = عدد الخطوات * المدى)",
-  "Batch count": "عدد الحزم",
-  "How many batches of images to create": "يتم إنتاج الصور على دفعات، كل دفعة فيها حزمة من الصور",
-  "Batch size": "حجم الحزمة",
-  "How many image to create in a single batch": "Batch size: إنتاج حزمة صور أسرع من إنتاجهم فرادى، حدد عدد الصور في كل حزمة",
-  "CFG Scale": "التركيز",
-  "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "CFG scale: يحدد مقدار التركيز على تلبية الطلب وتجنب عكسه، كلما زاد قل الإبداع",
-  "Seed": "البذرة",
-  "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Seed: رقم طبيعي عشوائي يسمح بإعادة إنتاج نفس الصورة إذا توافقت قيم العوامل الأخرى",
-  "Set seed to -1, which will cause a new random number to be used every time": "استخدم بذرة جديدة في كل مرة (نرمز لهذا الخيار بجعل قيمة البذرة 1-)",
-  "Reuse seed from last generation, mostly useful if it was randomed": "أعد استخدام البذرة من الإنتاج السابق",
-  "Extra": "مزج",
-  "Variation seed": "بذرة الممزوج",
-  "Seed of a different picture to be mixed into the generation.": "Variation seed: بذرة صورة أخرى ليتم مزجها مع الصورة الحالية",
-  "Variation strength": "أثر الممزوج",
-  "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Variation seed strength: مقدار أثر الصورة المدمجة على النتيجة النهائية (0: لا أثر، 1: أثر كامل ما عدا عند استخدام أسلوب خطو سلفي Ancestral)",
-  "Resize seed from width": "عرض الممزوج",
-  "Resize seed from height": "إرتفاع الممزوج",
-  "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Seed resize from: حدد دقة صورة الممزوج (0: نفس دقة الإنتاج)",
-  "Open for Clip Aesthetic!": "تضمين تجميلي",
-  "▼": "▼",
-  "Aesthetic weight": "أثر التضمين",
-  "Aesthetic steps": "عدد الخطوات",
-  "Aesthetic learning rate": "معدل التعلم",
-  "Slerp interpolation": "امزج بطريقة كروية",
-  "Aesthetic imgs embedding": "التضمين",
-  "None": "بدون",
-  "Aesthetic text for imgs": "الطلب (اختياري)",
-  "This text is used to rotate the feature space of the imgs embs": "لإعادة توجيه التضمين التجميلي",
-  "Slerp angle": "أثر الطلب",
-  "Is negative text": "الطلب عكسي",
-  "Script": "أدوات خاصة",
-  "Prompt matrix": "مصفوفة طلبات",
-  "Put variable parts at start of prompt": "الجزء المتغير في بداية الطلب",
-  "Prompts from file or textbox": " قائمة طلبات",
-  "Iterate seed every line": "غير البذرة مع كل طلب",
-  "List of prompt inputs": "قائمة الطلبات",
-  "Upload prompt inputs": "اجلب الطلبات من ملف",
-  "Drop File Here": "اسقط ملف هنا",
-  "-": "-",
-  "or": "أو",
-  "Click to Upload": "انقر للرفع",
-  "X/Y plot": "مصفوفة عوامل",
-  "X type": "العامل الأول",
-  "Nothing": "لا شيء",
-  "Var. seed": "بذرة الممزوج",
-  "Var. strength": "أثر الممزوج",
-  "Steps": "عدد الخطوات",
-  "Prompt S/R": "كلمات بديلة",
-  "Prompt order": "ترتيب الكلمات",
-  "Sampler": "أسلوب الخطو",
-  "Checkpoint name": "ملف الأوزان",
-  "Hypernetwork": "الشبكة الفائقة",
-  "Hypernet str.": "قوة الشبكة الفائقة",
-  "Inpainting conditioning mask strength": "قوة قناع الإنتاج الجزئي",
-  "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "حدد مدى صرامة قناع الإنتاج، يصبح القناع شفاف إذا قوته 0 (لا يعمل إلا مع ملفات أوزان الإنتاج الجزئي: inpainting)",
-  "Sigma Churn": "العشوائية (Schurn)",
-  "Sigma min": "أدنى تشويش (Stmin)",
-  "Sigma max": "أقصى تشويش (Stmax)",
-  "Sigma noise": "التشويش (Snoise)",
-  "Eta": "العامل Eta η",
-  "Clip skip": "تخطي آخر طبقات CLIP",
-  "Denoising": "المدى",
-  "Cond. Image Mask Weight": "قوة قناع الإنتاج الجزئي",
-  "X values": "قيم العامل الأول",
-  "Separate values for X axis using commas.": "افصل القيم بفواصل (,) من اليسار إلى اليمين",
-  "Y type": "العامل الثاني",
-  "Y values": "قيم العامل الثاني",
-  "Separate values for Y axis using commas.": "افصل القيم بفواصل (,) من الأعلى إلى الأسفل",
-  "Draw legend": "أضف مفتاح التوضيح",
-  "Include Separate Images": "أضف الصور منفصلة",
-  "Keep -1 for seeds": "استخدم بذور عشوائية",
-  "Save": "احفظ",
-  "Write image to a directory (default - log/images) and generation parameters into csv file.": "احفظ الصور مع ملف العوامل بصيغة CSV",
-  "Send to img2img": "أرسل لصورة إلى صورة",
-  "Send to inpaint": "أرسل للإنتاج الجزئي",
-  "Send to extras": "أرسل للمعالجة",
-  "Open images output directory": "افتح مجلد الصور المخرجة",
-  "Make Zip when Save?": "ضع النتائج في ملف مضغوط عند الحفظ",
-  "img2img": "صورة إلى صورة",
-  "Interrogate\nCLIP": "استجواب\nCLIP",
-  "Drop Image Here": "اسقط صورة هنا",
-  "Just resize": "تغيير الحجم فقط",
-  "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "غير حجم الصورة بدون مراعات اتزان الأبعاد",
-  "Crop and resize": "تغيير الحجم وقص الأطراف",
-  "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "غير حجم الصورة واقتص الأطراف الزائدة",
-  "Resize and fill": "تغيير الحجم وتبطين الأطراف",
-  "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "غير حجم الصورة واملأ الأطراف الزائدة بألوان من الصورة",
-  "img2img alternative test": "استجواب الصورة (تجريبي)",
-  "should be 2 or lower.": "يفترض أن يكون 2 أو أقل",
-  "Override `Sampling method` to Euler?(this method is built for it)": "استخدم أسلوب خطو Euler (مستحسن)",
-  "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "استبدل الطلب وعكسه في الأعلى بالطلب الأصلي وعكسه التاليين",
-  "Original prompt": "الطلب الأصلي",
-  "Original negative prompt": "عكس الطلب الأصلي",
-  "Override `Sampling Steps` to the same value as `Decode steps`?": "استبدل عدد الخطوات بعدد الخطوات الأصلية",
-  "Decode steps": "عدد الخطوات الأصلية",
-  "Override `Denoising strength` to 1?": "اجعل المدى 1",
-  "Decode CFG scale": "التركيز",
-  "Randomness": "العشوائية",
-  "Sigma adjustment for finding noise for image": "لا تسمح بتثبيت قيمة التباين",
-  "Loopback": "اجترار وتكرار",
-  "Loops": "عدد المرات",
-  "How many times to repeat processing an image and using it as input for the next iteration": "كم مرة يتم أخذ مخرجات الإنتاج كمدخلات وإعادة الإنتاج مرة أخرى",
-  "Denoising strength change factor": "معدل تغيير المدى",
-  "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "يتم ضرب المدى بهذا الرقم في كل مرة، إذا استخدمت رقم أصغر من 1 يمكن الرسو على نتيجة، وإذا استخدمت رقم أكبر من 1 تصبح النتيجة عشوائية",
-  "Outpainting mk2": "توسيع الصورة (mk2)",
-  "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "يفضل استخدام: 80-100 خطوة، أسلوب Euler a، المدى 0.8",
-  "Pixels to expand": "عدد البيكسلات",
-  "Mask blur": "تنعيم القناع",
-  "How much to blur the mask before processing, in pixels.": "مقدرا تنعيم القناع قبل استخدامه (يقاس بالبيكسل)",
-  "Outpainting direction": "اتجاه توسيع الصورة",
-  "left": "يسار",
-  "right": "يمين",
-  "up": "فوق",
-  "down": "تحت",
-  "Fall-off exponent (lower=higher detail)": "قوة السقوط (كلما قلت زادت التفاصيل)",
-  "Color variation": "تنوع الألوان",
-  "Poor man's outpainting": "توسيع الصورة (بدائي)",
-  "Masked content": "محتويات القناع",
-  "What to put inside the masked area before processing it with Stable Diffusion.": "ما يوضع مكان الفراغ في الصورة الذي نريد إنتاج محتوياته",
-  "fill": "ألوان",
-  "fill it with colors of the image": "املأ باستخدام ألوان مأخوذة من باقي الصورة",
-  "original": "بدون تغيير",
-  "keep whatever was there originally": "أبق محتويات ما تحت القناع كما هي",
-  "latent noise": "تشويش كامن",
-  "fill it with latent space noise": "املأه باستخدام تشويش من الفضاء الكامن",
-  "latent nothing": "تصفير كامن",
-  "fill it with latent space zeroes": "استبدل مكان القناع في الفضاء الكامن بأصفار",
-  "SD upscale": "مضاعفة الدقة بنموذج الإنتشار المستقر",
-  "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "سيتم تكبير حجم الصورة إلى الضعف، استخدم الطول والإرتفاع في الأعلى لتحديد حجم نافذة المكبر",
-  "Tile overlap": "تداخل النافذة",
-  "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "المكبر ينظر إلى أجزاء الصورة من خلال نافذة لتكبير المحتوى ثم ينتقل إلى الجزء المجاور، يفضل أن يكون هناك تداخل بين كل رقعة لكي لا يكون هناك اختلاف واضح بينهم",
-  "Upscaler": "طريقة التكبير",
-  "Lanczos": "Lanczos",
-  "LDSR": "LDSR",
-  "ScuNET GAN": "ScuNET GAN",
-  "ScuNET PSNR": "ScuNET PSNR",
-  "ESRGAN_4x": "ESRGAN_4x",
-  "SwinIR 4x": "SwinIR 4x",
-  "Inpaint": "إنتاج جزئي",
-  "Draw mask": "ارسم القناع",
-  "Upload mask": "ارفع القناع",
-  "Inpaint masked": "أنتج ما بداخل القناع",
-  "Inpaint not masked": "أنتج ما حول القناع",
-  "Inpaint at full resolution": "إنتاج بالدقة الكاملة",
-  "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "كبر ما يراد إعادة إنتاجه ثم صغر النتيجة وألصقها في مكانها",
-  "Inpaint at full resolution padding, pixels": "عدد بيكسلات التبطين",
-  "Batch img2img": "صور إلى صور",
-  "Process images in a directory on the same machine where the server is running.": "حدد مسار مجلد صور موجود في جهاز الخادم",
-  "Use an empty output directory to save pictures normally instead of writing to the output directory.": "يمكنك أيضا تحديد مجلد حفظ النتائج (غير الإفتراضي)",
-  "Input directory": "مجلد المدخلات",
-  "Output directory": "مجلد المخرجات",
-  "Extras": "معالجة",
-  "Single Image": "صورة واحدة",
-  "Source": "المصدر",
-  "Scale by": "مضاعفة الدقة",
-  "Resize": "تغيير الحجم",
-  "Upscaler 1": "المكبر الأول",
-  "Upscaler 2": "المكبر الثاني",
-  "Upscaler 2 visibility": "أثر المكبر الثاني",
-  "GFPGAN visibility": "أثر GFPGAN (محسن وجوه)",
-  "CodeFormer visibility": "أثر CodeFormer (محسن وجوه)",
-  "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "وزن CodeFormer (يزيد التفاصيل على حساب الجودة)",
-  "Upscale Before Restoring Faces": "كبر قبل تحسين الوجوه",
-  "Scale to": "دقة محددة",
-  "Crop to fit": "قص الأطراف الزائدة إذا لم تتناسب الأبعاد",
-  "Batch Process": "حزمة صور",
-  "Batch from Directory": "حزمة من مجلد",
-  "A directory on the same machine where the server is running.": "مسار مجلد صور موجود في جهاز الخادم",
-  "Leave blank to save images to the default path.": "اتركه فارغا لاستخدام المسار الإفتراضي",
-  "Show result images": "اعرض الصور الناتجة",
-  "PNG Info": "عوامل الصورة",
-  "Send to txt2img": "أرسل لنص إلى صورة",
-  "Checkpoint Merger": "مزج الأوزان",
-  "A merger of the two checkpoints will be generated in your": "سيتم مزج الأوزان التالية وحفظ الأوزان المدجمة مع ",
-  "checkpoint": "الأوزان",
-  "directory.": " مجلد.",
-  "Primary model (A)": "الأوزان الأولى (A)",
-  "Secondary model (B)": "الأوزان الثانية (B)",
-  "Tertiary model (C)": "الأوزان الثالثة (C)",
-  "Custom Name (Optional)": "الاسم الجديد (اختياري)",
-  "Multiplier (M) - set to 0 to get model A": "العامل M: مسافة الإبتعاد عن الأوزان الأولى A",
-  "Interpolation Method": "طريقة المزج",
-  "Weighted sum": "خطية",
-  "Result = A * (1 - M) + B * M": "النتيجة = A * (1 - M) + B * M",
-  "Add difference": "جمع الفرق",
-  "Result = A + (B - C) * M": "النتيجة = A + (B - C) * M",
-  "Save as float16": "احفظ بدقة float16",
-  "Run": "تشغيل",
-  "Train": "تدريب",
-  "See": "اقرأ",
-  "wiki": " الـwiki ",
-  "for detailed explanation.": "لمعرفة المزيد",
-  "Create embedding": "إنشاء تضمين",
-  "Name": "الاسم",
-  "Initialization text": "النص المبدأي",
-  "Number of vectors per token": "عدد المتجهات لكل وحدة لغوية",
-  "Overwrite Old Embedding": "استبدل التضمين القديم",
-  "Create hypernetwork": "إنشاء شبكة فائقة",
-  "Modules": "الأجزاء",
-  "Enter hypernetwork layer structure": "ترتيب مضاعفات عرض الطبقات",
-  "1st and last digit must be 1. ex:'1, 2, 1'": "المضاعفين الأول والأخير يجب أن يكونا 1، مثال: 1, 2, 1",
-  "Select activation function of hypernetwork": "دالة التنشيط",
-  "linear": "linear",
-  "relu": "relu",
-  "leakyrelu": "leakyrelu",
-  "elu": "elu",
-  "swish": "swish",
-  "tanh": "tanh",
-  "sigmoid": "sigmoid",
-  "celu": "celu",
-  "gelu": "gelu",
-  "glu": "glu",
-  "hardshrink": "hardshrink",
-  "hardsigmoid": "hardsigmoid",
-  "hardtanh": "hardtanh",
-  "logsigmoid": "logsigmoid",
-  "logsoftmax": "logsoftmax",
-  "mish": "mish",
-  "prelu": "prelu",
-  "rrelu": "rrelu",
-  "relu6": "relu6",
-  "selu": "selu",
-  "silu": "silu",
-  "softmax": "softmax",
-  "softmax2d": "softmax2d",
-  "softmin": "softmin",
-  "softplus": "softplus",
-  "softshrink": "softshrink",
-  "softsign": "softsign",
-  "tanhshrink": "tanhshrink",
-  "threshold": "threshold",
-  "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "تهيئة الأوزان (استخدم Kaiming مع relu وأمثالها وXavier مع sigmoid وأمثالها)",
-  "Normal": "Normal",
-  "KaimingUniform": "KaimingUniform",
-  "KaimingNormal": "KaimingNormal",
-  "XavierUniform": "XavierUniform",
-  "XavierNormal": "XavierNormal",
-  "Add layer normalization": "أضف تسوية الطبقات (LayerNorm)",
-  "Use dropout": "استخدم الإسقاط (Dropout)",
-  "Overwrite Old Hypernetwork": "استبدل الشبكة الفائقة القديمة",
-  "Preprocess images": "معالجة مسبقة للصور",
-  "Source directory": "مجلد المدخلات",
-  "Destination directory": "مجلد المخرجات",
-  "Existing Caption txt Action": "اذا كانت الصورة لديها توصيف (طلب)",
-  "ignore": "تجاهل",
-  "copy": "انسخ",
-  "prepend": "أسبق",
-  "append": "ألحق",
-  "Create flipped copies": "انشئ نسخ معكوسة للصور",
-  "Split oversized images": "قسّم الصور الكبيرة",
-  "Split image threshold": "حد تقسيم الصور الكبيرة",
-  "Split image overlap ratio": "نسبة تداخل اقسام الصور الكبيرة",
-  "Auto focal point crop": "اقتصاص تلقائي",
-  "Focal point face weight": "تمركز الوجوه",
-  "Focal point entropy weight": "تمركز التنوع",
-  "Focal point edges weight": "تمركز الحواف",
-  "Create debug image": "احفظ نتائج التحليل أيضا",
-  "Use BLIP for caption": "استخدم BLIP لتوصيف الصور",
-  "Use deepbooru for caption": "استخدم deepbooru لتوصيف الصور",
-  "Preprocess": "معالجة مسبقة",
-  "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "درب التضمين أو الشبكة الفائقة: يجب تحديد مجلد يحتوي صور مربعة فقط ",
-  "[wiki]": "[wiki]",
-  "Embedding": "التضمين",
-  "Embedding Learning rate": "معدل تعلم التضمين",
-  "Hypernetwork Learning rate": "معدل تعلم الشبكة الفائقة",
-  "Dataset directory": "مجلد مجموعة البيانات",
-  "Path to directory with input images": "مسار مجلد الصور المدخلة",
-  "Log directory": "مجلد السجل",
-  "Path to directory where to write outputs": "مسار مجلد الصور المخرجة",
-  "Prompt template file": "ملف صيغ الطلبات",
-  "Max steps": "أقصى عدد لخطوات التدريب",
-  "Save an image to log directory every N steps, 0 to disable": "احفظ صورة في السجل بعد كل كم خطوة تدريب (إذا 0 لا تحفظ)",
-  "Save a copy of embedding to log directory every N steps, 0 to disable": "احفظ التضمين في السجل بعد كل كم خطوة تدريب (إذا 0 لا تحفظ)",
-  "Save images with embedding in PNG chunks": "احفظ التضمين بداخل ملف الصورة كعامل يمكن استخراجه من عوامل الصورة (صيغة PNG)",
-  "Read parameters (prompt, etc...) from txt2img tab when making previews": "استخدم قيم العوامل الموجودة في تبويب نص إلى صورة لعرض نتائجهم أثناء التدريب",
-  "Train Hypernetwork": "درّب الشبكة الفائقة",
-  "Train Embedding": "درّب التضمين",
-  "Create aesthetic embedding": "تضمين تجميلي",
-  "Create an aesthetic embedding out of any number of images": "انشئ تضمين تجميلي يعبر عن مجموعة من الصور",
-  "Create images embedding": "انشئ التضمين التجميلي",
-  "Image Browser": "معرض الصور",
-  "Load": "حمّل",
-  "Images directory": "مجلد الصور",
-  "First Page": "الصفحة الأولى",
-  "Prev Page": "الصفحة السابقة",
-  "Page Index": "رقم الصفحة",
-  "Next Page": "الصفحة التالية",
-  "End Page": "الصفحة الأخيرة",
-  "number of images to delete consecutively next": "عدد الصور المتتالية للحذف",
-  "Delete": "احذف",
-  "Generate Info": "معلومات عامة",
-  "File Name": "اسم الملف",
-  "Collect": "اجمع",
-  "extras": "معالجة",
-  "favorites": "المفضلة",
-  "custom fold": "مجلد آخر",
-  "Input images directory": "مجلد الصور المدخلة",
-  "Settings": "إعدادات",
-  "Apply settings": "طبق الإعدادت",
-  "Saving images/grids": "حفظ الصور وجداولها",
-  "Always save all generated images": "احفظ كل الصور المنتجة",
-  "File format for images": "صيغة ملفات الصور",
-  "Images filename pattern": "نمط تسمية الصور",
-  "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "استخدم الأوسمة التالية لتحديد كيفية تسمية الصور: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp] أو اتركه فارغا إذا أردت",
-  "Add number to filename when saving": "دائما أضف رقم على اسم الملف",
-  "Always save all generated image grids": "احفظ جداول الصور دائما",
-  "File format for grids": "صيغة ملفات جداول الصور",
-  "Add extended info (seed, prompt) to filename when saving grid": "أضف عوامل الإنتاج (مثل الطلب والبذرة) لأسماء جداول الصور",
-  "Do not save grids consisting of one picture": "لا تحفظ جدول الصور عند إنتاج صورة واحدة فقط",
-  "Prevent empty spots in grid (when set to autodetect)": "في الوضع التلقائي امنع الفراغات في جداول الصور",
-  "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "عدد صفوف جداول الصور (1-: تلقائي، 0: نفس حجم الحزمة)",
-  "Save text information about generation parameters as chunks to png files": "احفظ عوامل الإنتاج داخل ملفات الصور (صيغة PNG)",
-  "Create a text file next to every image with generation parameters.": "انشئ ملف نصي يحتوي على عوامل الإنتاج بجانب كل صورة",
-  "Save a copy of image before doing face restoration.": "احفظ نسخة من الصورة قبل تحسين الوجوه",
-  "Quality for saved jpeg images": "جودة حفظ صور JPEG",
-  "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "إذا كان حجم ملف صورة PNG أكبر من 4MB أو أحد أبعاد الصورة أكبر من 4000 بيكسل، صغر الصورة واحفظها بصيغة JPEG",
-  "Use original name for output filename during batch process in extras tab": "استخدم الإسم الأصلي للصور عند معالجتهم في حزم تحت تبويب معالجة",
-  "When using 'Save' button, only save a single selected image": "احفظ صورة واحدة فقط عند الضغط على حفظ",
-  "Do not add watermark to images": "لا تضف علامة مائية على الصور",
-  "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "الصور المنتجة لن تحتوي على علامة مائية عند تفعيل هذا الخيار، تحذير: ربما قد يدل هذا على أنك تتصرف بشكل غير أخلاقي",
-  "Paths for saving": "اماكن الحفظ",
-  "Output directory for images; if empty, defaults to three directories below": "مسار حفظ الصور المخرجة؛ يمكن أن يترك فارغا",
-  "Output directory for txt2img images": "مسار الصور المخرجة من تبويب نص إلى صورة",
-  "Output directory for img2img images": "مسار الصور المخرجة من تبويب صورة إلى صورة",
-  "Output directory for images from extras tab": "مسار الصور المخرجة من تبويب معالجة",
-  "Output directory for grids; if empty, defaults to two directories below": "مسار حفظ جداول الصور المخرجة؛ يمكن أن يترك فارغا",
-  "Output directory for txt2img grids": "مسار جداول الصور المخرجة من تبويب نص إلى صورة",
-  "Output directory for img2img grids": "مسار جداول الصور المخرجة من تبويب صورة إلى صورة",
-  "Directory for saving images using the Save button": "مسار حفظ الصور عند الضغط على زر الحفظ",
-  "Saving to a directory": "مجلدات الحفظ",
-  "Save images to a subdirectory": "احفظ الصور في مجلد فرعي",
-  "Save grids to a subdirectory": "احفظ جداول الصور في مجلد فرعي",
-  "When using \"Save\" button, save images to a subdirectory": "احفظ الصور في مجلد فرعي عند الضغط على زر الحفظ",
-  "Directory name pattern": "نمط اسم المجلد",
-  "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "استخدم الأوسمة التالية لتحديد كيفية تسمية المجلدات الفرعية للصور والصور المركبة: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp] أو اتركه فارغا إذا أردت",
-  "Max prompt words for [prompt_words] pattern": "أقصى عدد كلمات الطلب عند استخدام وسم [prompt_words]",
-  "Upscaling": "تكبير الصور",
-  "Tile size for ESRGAN upscalers. 0 = no tiling.": "حجم نافذة المكبر ESRGAN (يمكن أن يكون 0)",
-  "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "تداخل نافذة المكبر ESRGAN (تتضح الفواصل إذا كان قليل)",
-  "Select which Real-ESRGAN models to show in the web UI. (Requires restart)": "حدد ما تريد عرضه كخيار لمكبر من نوع Real-ESRGAN يمكن استخدامه (يتطلب إعادة تشغيل)",
-  "Tile size for all SwinIR.": "حجم نافذة المكبر SwinIR",
-  "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "تداخل نافذة المكبر SwinIR (تتضح الفواصل إذا كان قليل)",
-  "LDSR processing steps. Lower = faster": "عدد خطوات مكبر LDSR (أسرع كلما قلت)",
-  "Upscaler for img2img": "طريقة التكبير تحت تبويب صورة إلى صورة",
-  "Upscale latent space image when doing hires. fix": "عند اختيار \"إصلاح الدقة العالية\" قم بالتكبير في الفضاء الكامن",
-  "Face restoration": "تحسين الوجوه",
-  "Face restoration model": "نموذج تحسين الوجوه",
-  "Restore low quality faces using GFPGAN neural network": "استخدم نموذج GFPGAN لتحسين الوجوه",
-  "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "وزن CodeFormer (يزيد التفاصيل على حساب الجودة)",
-  "Move face restoration model from VRAM into RAM after processing": "احذف محسن الوجوه من ذاكرة كرت الشاشة (VRAM) بعد استخدامه",
-  "System": "النظام",
-  "VRAM usage polls per second during generation. Set to 0 to disable.": "عدد مرات التحقق من ذاكرة كرت الشاشة المستخدمة (VRAM) في الثانية",
-  "Always print all generation info to standard output": "اطبع دائما كل عوامل الإنتاج",
-  "Add a second progress bar to the console that shows progress for an entire job.": "أضف شريط تقدم ثاني لعملية الإنتاج الكاملة",
-  "Training": "التدريب",
-  "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "قلل استخدام ذاكرة كرت الشاشة (VRAM) عند تدريب الشبكة الفائقة بالتخلص من CLIP وVAE",
-  "Filename word regex": "التعبير النمطي (RegEx) لكلمات اسم الملف",
-  "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "استخدم هذا التعبير لاستخراج كلمات من أسماء الملفات ليتم دمجها بالخيار التالي لتستخدم في التدريب، اتركه فارغا لتستخدم اسم الملف كما هو",
-  "Filename join string": "النص الفاصل للكلمات المدموجة",
-  "This string will be used to join split words into a single line if the option above is enabled.": "سيستخدم بين كل كلمة يتم استخراجها من الخيار السابق",
-  "Number of repeats for a single input image per epoch; used only for displaying epoch number": "عدد مرات تكرار الصور في كل دورة (Epoch)، يستخدم للعرض فقط",
-  "Save an csv containing the loss to log directory every N steps, 0 to disable": "احفظ خسارة التدريب (Loss) في السجل بعد كل كم خطوة (إذا 0 لا تحفظ)",
-  "Stable Diffusion": "نموذج الإنتشار المستقر (Stable Diffusion)",
-  "Checkpoints to cache in RAM": "عدد النماذج التي تترك في الذاكرة العشوائية (RAM)",
-  "Hypernetwork strength": "قوة الشبكة الفائقة",
-  "Apply color correction to img2img results to match original colors.": "صحح ألوان نتائج صورة إلى صورة لتشابه الصورة الأصلية",
-  "Save a copy of image before applying color correction to img2img results": "احفظ نسخة من الصور المنتجة من صورة إلى صورة قبل عملية تصحيح الألوان",
-  "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "استخدم بالضبط عدد الخطوات المحددة بالرغم عن المدى (عدد الخطوات الإفتراضي = عدد الخطوات / المدى)",
-  "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "فعل التجزيء الكمي (Quantization) لأغلب أساليب الخطو (k-diffusion) للحصول على صور أنظف (يتطلب إعادة تشغيل)",
-  "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "أحط أي كلمة أو عبارة في الطلب أو عكسه بأقواس () للتشديد أو [] للتيسير",
-  "Use old emphasis implementation. Can be useful to reproduce old seeds.": "استخدم الطريقة القديمة للتشديد (إذا كانت لديك نتائج قديمة تريد تجربتها)",
-  "Make K-diffusion samplers produce same images in a batch as when making a single image": "عدم تأثير تحزيم الصور على البذور عند استخدام أساليب خطو (k-diffusion)",
-  "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "عند استخدام طلب طويل أكبر من 75 وحدة لغوية، افصل الطلب عند أخر فاصلة قبل كم كلمة",
-  "Filter NSFW content": "انتق المحتوى النظيف فقط",
-  "Stop At last layers of CLIP model": "قف عند آخر كم طبقة لنموذج CLIP",
-  "Allowed categories for random artists selection when using the Roll button": "اختر اهتمامات الفنانين المسموح بإضافتهم للطلب",
-  "Interrogate Options": "خيارات الاستجواب",
-  "Interrogate: keep models in VRAM": "ابق نموذج الاستجواب في ذاكرة كرت الشاشة (VRAM)",
-  "Interrogate: use artists from artists.csv": "استخدم قائمة الفنانين في الاستجواب من ملف artists.csv",
-  "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "أدرج رتب الشعارات عند الاستجواب (لا تعمل مع جميع النماذج)",
-  "Interrogate: num_beams for BLIP": "عدد أشعة الاستجواب لنموذج BLIP",
-  "Interrogate: minimum description length (excluding artists, etc..)": "أقل عدد كلمات للتوصيف (لا يتضمن الفنانين وغيرهم)",
-  "Interrogate: maximum description length": "أكثر عدد كلمات للتوصيف",
-  "CLIP: maximum number of lines in text file (0 = No limit)": "أكثر عدد أسطر لملف نصي عند الاستجواب باستخدام CLIP (0 = بدون حد)",
-  "Interrogate: deepbooru score threshold": "حد درجة الاستجواب باستخدام deepbooru",
-  "Interrogate: deepbooru sort alphabetically": "رتب نتائج توصيف deepbooru أبجديا",
-  "use spaces for tags in deepbooru": "افصل شعارات deepbooru بمسافات",
-  "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "تجاهل الأقواس عند استخدام deepbooru (كي لا تعامل كأقواس التشديد)",
-  "User interface": "واجهة المستخدم",
-  "Show progressbar": "اظهر شريط التقدم",
-  "Show image creation progress every N sampling steps. Set 0 to disable.": "اعرض صورة بعد كل كم خطوة (إذا 0 لا تعرض)",
-  "Show previews of all images generated in a batch as a grid": "اعرض كل الصور التي تم إنتاجها في حزمة كجدول",
-  "Show grid in results for web": "أظهر نتائج جداول الصور",
-  "Do not show any images in results for web": "لا تظهر نتائج الصور",
-  "Add model hash to generation information": "أضف رمز تهشير (Hash) ملف الأوزان لعوامل الإنتاج",
-  "Add model name to generation information": "أضف اسم ملف الأوزان لعوامل الإنتاج",
-  "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "لا تغير الأوزان المختارة عند قراءة عوامل الإنتاج من صورة أو من ملف",
-  "Send seed when sending prompt or image to other interface": "عند إرسال صورة أو طلب ألحق البذرة أيضا",
-  "Font for image grids that have text": "نوع الخط في جداول الصور التي تحتوي على نصوص",
-  "Enable full page image viewer": "اسمح بعرض الصور في وضع ملئ الشاشة",
-  "Show images zoomed in by default in full page image viewer": "اعرض الصور مقربة عند استخدام وضع ملئ الشاشة",
-  "Show generation progress in window title.": "أظهر التقدم في عنوان النافذة",
-  "Quicksettings list": "قائمة الإعدادات السريعة",
-  "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "قائمة مقسمة بفواصل لأسماء الإعدادات التي يجب أن تظهر في الأعلى لتسهيل الوصول إليها، انظر إلى modules/shared.py لمعرفة الأسماء، يتطلب إعادة تشغيل",
-  "Localization (requires restart)": "اللغة (تتطلب إعادة تشغيل)",
-  "pt_BR": "البرتغالية",
-  "zh_CN": "الصينية",
-  "ko_KR": "الكورية",
-  "fr_FR": "الفرنسية",
-  "ru_RU": "الروسية",
-  "ar_AR": "العربية",
-  "tr_TR": "التركية",
-  "it_IT": "الإيطالية",
-  "ja_JP": "اليابانية",
-  "de_DE": "الألمانية",
-  "zh_TW": "الصينية (تايوان)",
-  "es_ES": "الإسبانية",
-  "Sampler parameters": "عوامل أساليب الخطو",
-  "Hide samplers in user interface (requires restart)": "اخف أساليب الخطو التالية (يتطلب إعادة تشغيل)",
-  "eta (noise multiplier) for DDIM": "العامل Eta η لأسلوب الخطو DDIM",
-  "eta (noise multiplier) for ancestral samplers": "العامل Eta η لأساليب الخطو السلفية (Ancestral)",
-  "img2img DDIM discretize": "طريقة التقطيع (Discretization) لأسلوب الخطو DDIM في وضع صورة إلى صورة",
-  "uniform": "خطية",
-  "quad": "تربيعية",
-  "sigma churn": "العشوائية (Schurn)",
-  "sigma tmin": "أدنى تشويش (Stmin)",
-  "sigma noise": "التشويش (Snoise)",
-  "Eta noise seed delta": "إزاحة بذرة أساليب الخطو التي تستعمل العامل Eta η",
-  "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "هذا الرقم سيتم إضافته إلى البذرة عند استخدام أحد أساليب الخطو التي تستعمل العامل Eta η، يفيد هذا الخيار في مشابهة نتائج بعض البرامج الأخرى التي تستعمله",
-  "Images Browser": "معرض الصور",
-  "Preload images at startup": "حمل الصور عند بدء التشغيل",
-  "Number of columns on the page": "عدد الأعمدة في كل صفحة",
-  "Number of rows on the page": "عدد الصفوف في كل صفحة",
-  "Minimum number of pages per load": "أقل عدد صور يتم تحميلها كل مرة",
-  "Request browser notifications": "اطلب تنبيهات المتصفح",
-  "Download localization template": "حمل ملف اللغة",
-  "Reload custom script bodies (No ui updates, No restart)": "أعد تحميل الأدوات الخاصة (بدون واجهة المستخدم ولا يحتاج إعادة تشغيل)",
-  "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "أعد تشغيل gradio وتحميل الأدوات الخاصة وواجهة المستخدم",
-  "⤡": "⤡",
-  "⊞": "⊞",
-  "×": "×",
-  "❮": "❮",
-  "❯": "❯",
-  "•": "•",
-  "Label": "Label",
-  "File": "File",
-  "Image": "Image",
-  "Check progress": "Check progress",
-  "Check progress (first)": "Check progress (first)",
-  "Textbox": "Textbox",
-  "Image for img2img": "Image for img2img",
-  "Image for inpainting with mask": "Image for inpainting with mask",
-  "Mask": "Mask",
-  "Mask mode": "Mask mode",
-  "Masking mode": "Masking mode",
-  "Resize mode": "Resize mode",
-  "Prev batch": "Prev batch",
-  "Next batch": "Next batch",
-  "Refresh page": "Refresh page",
-  "Date to": "Date to",
-  "Number": "Number",
-  "set_index": "set_index",
-  "Checkbox": "Checkbox"
-}

localizations/de_DE.json

@@ -1,458 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "view": "API ",
-    "api": "anzeigen",
-    "•": " • ",
-    "built with gradio": "Mit Gradio erstellt",
-    "Loading...": "Lädt...",
-    "Stable Diffusion checkpoint": "Stable Diffusion Checkpoint",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "Extras",
-    "PNG Info": "PNG Info",
-    "Checkpoint Merger": "Checkpoint Fusion",
-    "Train": "Trainieren",
-    "Settings": "Einstellungen",
-    "Prompt": "Prompt",
-    "Negative prompt": "Negative Prompt",
-    "Run": "Ausführen",
-    "Skip": "Überspringen",
-    "Interrupt": "Abbrechen",
-    "Generate": "Generieren",
-    "Style 1": "Stil 1",
-    "Style 2": "Stil 2",
-    "Label": "Bezeichnung",
-    "File": "Datei",
-    "Drop File Here": "Datei hier ablegen",
-    "-": "-",
-    "o": "oder",
-    "Click to Upload": "Hochladen",
-    "Image": "Bild",
-    "Check progress": "Fortschitt prüfen",
-    "Check progress (first)": "Fortschritt prüfen (Initial)",
-    "Sampling Steps": "Samplingschritte",
-    "Sampling method": "Samplingmethode",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Breite",
-    "Height": "Höhe",
-    "Restore faces": "Gesichter wiederherstellen",
-    "Tiling": "Kacheln",
-    "Highres. fix": "Highres. Fix",
-    "Firstpass width": "Breite Erstdurchlauf",
-    "Firstpass height": "Höhe Erstdurchlauf",
-    "Denoising strength": "Denoisingstärke",
-    "Batch count": "Batchanzahl",
-    "Batch size": "Batchgröße",
-    "CFG Scale": "CFG-Skala",
-    "Seed": "Seed",
-    "Extra": "Extra",
-    "Variation seed": "Variationsseed",
-    "Variation strength": "Variationsstärke",
-    "Resize seed from width": "Seed von Breite ändern",
-    "Resize seed from height": "Seed von Höhe ändern",
-    "Script": "Skript",
-    "None": "Nichts",
-    "Prompt matrix": "Promptmatrix",
-    "Prompts from file or textbox": "Prompts aus Datei oder Textfeld",
-    "X/Y plot": "X/Y Graph",
-    "Put variable parts at start of prompt": "Variable teile am start des Prompt setzen",
-    "Iterate seed every line": "Iterate seed every line",
-    "List of prompt inputs": "List of prompt inputs",
-    "Upload prompt inputs": "Upload prompt inputs",
-    "X type": "X-Typ",
-    "Nothing": "Nichts",
-    "Var. seed": "Var. seed",
-    "Var. strength": "Var. strength",
-    "Steps": "Schritte",
-    "Prompt S/R": "Prompt Suchen/Ersetzen",
-    "Prompt order": "Promptreihenfolge",
-    "Sampler": "Sampler",
-    "Checkpoint name": "Checkpointname",
-    "Hypernetwork": "Hypernetwork",
-    "Hypernet str.": "Hypernet str.",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma min",
-    "Sigma max": "Sigma max",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Eta",
-    "Clip skip": "Clip skip",
-    "Denoising": "Denoising",
-    "Cond. Image Mask Weight": "Cond. Image Mask Weight",
-    "X values": "X-Werte",
-    "Y type": "Y-Typ",
-    "Y values": "Y-Werte",
-    "Draw legend": "Legende zeichnen",
-    "Include Separate Images": "Seperate Bilder hinzufügen",
-    "Keep -1 for seeds": "-1 als Seed behalten",
-    "Save": "Speichern",
-    "Send to img2img": "An img2img senden",
-    "Send to inpaint": "An Inpaint senden",
-    "Send to extras": "An Extras senden",
-    "Make Zip when Save?": "Zip beim Speichern erstellen?",
-    "Textbox": "Textfeld",
-    "Interrogate\nCLIP": "Interrogate\nCLIP",
-    "Interrogate\nDeepBooru": "Interrogate\nDeepBooru",
-    "Inpaint": "Inpaint",
-    "Batch img2img": "Batch img2img",
-    "Image for img2img": "Bild für img2img",
-    "Drop Image Here": "Bild hier ablegen",
-    "Image for inpainting with mask": "Bild für inpainting mit Maske",
-    "Mask": "Maske",
-    "Mask blur": "Maskenunschärfe",
-    "Mask mode": "Maskenmodus",
-    "Draw mask": "Maske zeichnen",
-    "Upload mask": "Maske hochladen",
-    "Masking mode": "Maskierungsmodus",
-    "Inpaint masked": "Maskiertes inpainten",
-    "Inpaint not masked": "Nicht maskiertes inpainten",
-    "Masked content": "Maskierter Inhalt",
-    "fill": "ausfüllen",
-    "original": "original",
-    "latent noise": "latent noise",
-    "latent nothing": "latent nothing",
-    "Inpaint at full resolution": "Inpaint mit voller Auflösung",
-    "Inpaint at full resolution padding, pixels": "Inpaint bei voller Auflösung Abstand, Pixel",
-    "Process images in a directory on the same machine where the server is running.": "Bilder in einem Verzeichnis auf demselben Rechner verarbeiten, auf dem der Server läuft.",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Ein leeres Ausgabeverzeichnis verwenden, um Bilder normal zu speichern, anstatt in das Ausgabeverzeichnis zu schreiben.",
-    "Input directory": "Eingabeverzeichnis",
-    "Output directory": "Ausgabeverzeichnis",
-    "Resize mode": "Größenänderungsmodus",
-    "Just resize": "Nur Größe anpassen",
-    "Crop and resize": "Zuschneiden und Größe anpassen",
-    "Resize and fill": "Größe anpassen und ausfüllen",
-    "img2img alternative test": "img2img alternativer Test",
-    "Loopback": "Loopback",
-    "Outpainting mk2": "Outpainting mk2",
-    "Poor man's outpainting": "Poor man's outpainting",
-    "SD upscale": "SD-Upscale",
-    "should be 2 or lower.": "Sollte 2 oder niedriger sein.",
-    "Override `Sampling method` to Euler?(this method is built for it)": "`Samplingmethode` auf Euler setzen? (Diese Methode is dafür ausgelegt)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "`Prompt` auf denselben Wert wie `Originale Prompt` (und `Negative Prompt`) setzen?",
-    "Original prompt": "Originale Prompt",
-    "Original negative prompt": "Originale negative Prompt",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "`Samplingschritte` auf denselben Wert wie `Dekodierschritte` setzen?",
-    "Decode steps": "Dekodierschritte",
-    "Override `Denoising strength` to 1?": "`Denoisingstärke auf 1 setzen?",
-    "Decode CFG scale": "CFG-Skala dekodieren",
-    "Randomness": "Zufälligkeit",
-    "Sigma adjustment for finding noise for image": "Sigma-Anpassung für die Suche nach Noise des Bildes",
-    "Loops": "Schleifen",
-    "Denoising strength change factor": "Denoising strength change factor",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Empfohlene Einstellungen: Samplingschritte: 80-100, Samplermethode: Euler a, Denoisingstärke: 0.8",
-    "Pixels to expand": "Pixel zum Erweitern",
-    "Outpainting direction": "Outpainting Richtung",
-    "left": "Links",
-    "right": "Rechts",
-    "up": "Hoch",
-    "down": "Runter",
-    "Fall-off exponent (lower=higher detail)": "Abfallexponent (niedriger=mehr Details)",
-    "Color variation": "Farbabweichung",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Skaliert das Bild auf die doppelte Größe; Benutze die Schieberegler für Breite und Höhe, um die Kachelgröße einzustellen",
-    "Tile overlap": "Kachelüberlappung",
-    "Upscaler": "Upscaler",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "SwinIR 4x": "SwinIR 4x",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "Single Image": "Einzelnes Bild",
-    "Batch Process": "Batchverarbeitung",
-    "Batch from Directory": "Batchverarbeitung aus Verzeichnis",
-    "Source": "Quelle",
-    "Show result images": "Bildergebnisse zeigen",
-    "Scale by": "Skalieren um",
-    "Scale to": "Skalieren zu",
-    "Resize": "Größe anpassen",
-    "Crop to fit": "Zuschneiden damit es passt",
-    "Upscaler 2 visibility": "Upscaler 2 Sichtbarkeit",
-    "GFPGAN visibility": "GFPGAN Sichtbarkeit",
-    "CodeFormer visibility": "CodeFormer Sichtbarkeit",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormer Gewichtung (0 = maximale Wirkung, 1 = minimale Wirkung)",
-    "Upscale Before Restoring Faces": "Upscale Before Restoring Faces",
-    "Send to txt2img": "An txt2img senden",
-    "A merger of the two checkpoints will be generated in your": "Die zusammgeführten Checkpoints werden gespeichert unter",
-    "checkpoint": "Checkpoint",
-    "directory.": "Verzeichnis.",
-    "Primary model (A)": "Primäres Modell (A)",
-    "Secondary model (B)": "Sekundäres Modell (B)",
-    "Tertiary model (C)": "Tertiäres Modell (C)",
-    "Custom Name (Optional)": "Eigener Name (Optional)",
-    "Multiplier (M) - set to 0 to get model A": "Multiplikator (M) - auf 0 setzen, um Modell A zu erhalten",
-    "Interpolation Method": "Interpolationsmethode",
-    "Weighted sum": "Weighted sum",
-    "Add difference": "Add difference",
-    "Save as float16": "Speichern als float16",
-    "See": "Siehe ",
-    "wiki": "Wiki ",
-    "for detailed explanation.": "für eine ausführliche Erklärung.",
-    "Create embedding": "Embedding erstellen",
-    "Create hypernetwork": "Hypernetwork erstellen",
-    "Preprocess images": "Bilder vorbereiten",
-    "Name": "Name",
-    "Initialization text": "Initialisierungstext",
-    "Number of vectors per token": "Anzahl der Vektoren pro Token",
-    "Overwrite Old Embedding": "Alte Embeddings überschreiben",
-    "Modules": "Module",
-    "Enter hypernetwork layer structure": "Hypernetwork-Ebenenstruktur angeben",
-    "Select activation function of hypernetwork": "Aktivierungsfunktion des Hypernetwork auswählen",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "tanh": "tanh",
-    "sigmoid": "sigmoid",
-    "celu": "celu",
-    "gelu": "gelu",
-    "glu": "glu",
-    "hardshrink": "hardshrink",
-    "hardsigmoid": "hardsigmoid",
-    "hardtanh": "hardtanh",
-    "logsigmoid": "logsigmoid",
-    "logsoftmax": "logsoftmax",
-    "mish": "mish",
-    "prelu": "prelu",
-    "rrelu": "rrelu",
-    "relu6": "relu6",
-    "selu": "selu",
-    "silu": "silu",
-    "softmax": "softmax",
-    "softmax2d": "softmax2d",
-    "softmin": "softmin",
-    "softplus": "softplus",
-    "softshrink": "softshrink",
-    "softsign": "softsign",
-    "tanhshrink": "tanhshrink",
-    "threshold": "threshold",
-    "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "Auswahl der Initialisierung der Ebenengewichte. Empfohlen wird relu-like - Kaiming, sigmoid-like - Xavier",
-    "Normal": "Normal",
-    "KaimingUniform": "KaimingUniform",
-    "KaimingNormal": "KaimingNormal",
-    "XavierUniform": "XavierUniform",
-    "XavierNormal": "XavierNormal",
-    "Add layer normalization": "Ebenennormalisierung hinzufügen",
-    "Use dropout": "Dropout benutzen",
-    "Overwrite Old Hypernetwork": "Altes Hypernetwork überschreiben",
-    "Source directory": "Quellenverzeichnis",
-    "Destination directory": "Zielverzeichnis",
-    "Existing Caption txt Action": "Vorhandene Beschriftung der txt",
-    "ignore": "ignorieren",
-    "copy": "kopieren",
-    "prepend": "voranstellen",
-    "append": "anhängen",
-    "Create flipped copies": "Gespiegelte Bilder erstellen",
-    "Split oversized images": "Übergroße Bilder aufteilen",
-    "Auto focal point crop": "Automatisch auf Fokuspunkt zuschneiden",
-    "Use BLIP for caption": "BLIP für Beschriftung nutzen",
-    "Use deepbooru for caption": "Deepbooru für Beschriftung nutzen",
-    "Split image threshold": "Schwellenwert für die Aufteilung von Bildern",
-    "Split image overlap ratio": "Überschneidungsverhältnis der Teilbilder",
-    "Focal point face weight": "Fokuspunkt Gesicht Gewicht",
-    "Focal point entropy weight": "Fokuspunkt Entropie Gewicht",
-    "Focal point edges weight": "Fokuspunkt Kanten Gewicht",
-    "Create debug image": "Testbild erstellen",
-    "Preprocess": "Vorbereiten",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "Trainieren eines Embeddings oder eines Hypernetworks; Sie müssen ein Verzeichnis mit einem Satz von Bildern im Verhältnis 1:1 angeben",
-    "[wiki]": "[Wiki]",
-    "Embedding": "Embedding",
-    "Embedding Learning rate": "Embedding Lernrate",
-    "Hypernetwork Learning rate": "Hypernetwork Lernrate",
-    "Dataset directory": "Datensatzverzeichnis",
-    "Log directory": "Protokollverzeichnis",
-    "Prompt template file": "Prompt-Vorlagendatei",
-    "Max steps": "Max Schritte",
-    "Save an image to log directory every N steps, 0 to disable": "Speichere alle N Schritte ein Bild im Protokollverzeichnis, 0 zum Deaktivieren",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Speichere alle N Schritte eine Embeddingkopie im Protokollverzeichnis, 0 zum Deaktivieren",
-    "Save images with embedding in PNG chunks": "Speichere Bilder mit Embeddings in PNG Chunks",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Lese Parameter (Prompt, etc...) aus dem txt2img-Tab beim Erstellen von Vorschaubildern.",
-    "Train Hypernetwork": "Hypernetwork Trainieren",
-    "Train Embedding": "Embedding Trainieren",
-    "Apply settings": "Eintellungen anwenden",
-    "Saving images/grids": "Bilder/Raster speichern",
-    "Always save all generated images": "Immer alle generierten Bilder speichern",
-    "File format for images": "Dateiformat für Bilder",
-    "Images filename pattern": "Dateinamensmuster für Bilder",
-    "Add number to filename when saving": "Beim speichern, dem Dateinamen Nummer anhängen",
-    "Always save all generated image grids": "Immer alle generierten Bildraster speichern",
-    "File format for grids": "Dateiformat für Raster",
-    "Add extended info (seed, prompt) to filename when saving grid": "Beim Speichern von Rastern zusätzliche Information (Seed, Prompt) hinzufügen",
-    "Do not save grids consisting of one picture": "Keine Raster speichern, die nur aus einem Bild bestehen",
-    "Prevent empty spots in grid (when set to autodetect)": "Lücken im Raster verhindern (falls auf Auto-Erkennung gesetzt)",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Rasterreihenanzahl; -1 für Auto-Erkennung und 0 für die gleiche wie die Batchanzahl",
-    "Save text information about generation parameters as chunks to png files": "Generationsparameter als Chunks in PNG-Dateien speichern",
-    "Create a text file next to every image with generation parameters.": "Erstelle zu jedem Bild eine Textdatei, die die Generationsparameter enthält",
-    "Save a copy of image before doing face restoration.": "Vor der Gesichtswiederhestellung eine Kopie des Bildes speichern",
-    "Quality for saved jpeg images": "Qualität der JPEG-Bilder",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "Wenn ein PNG-Bild größer als 4MB oder die Dimensionen größer als 4000 ist, herunterskalieren und als JPG speichern.",
-    "Use original name for output filename during batch process in extras tab": "Orginale Dateinamen als Ausgabenamen bei der Batchverarbeitung im Extras-Tab verwenden",
-    "When using 'Save' button, only save a single selected image": "Bei der Benutzung des 'Speichern'-Knopfes, nur das ausgewählte Bild speichern",
-    "Do not add watermark to images": "Den Bildern kein Wasserzeichen hinzufügen",
-    "Paths for saving": "Pfade zum Speichern",
-    "Output directory for images; if empty, defaults to three directories below": "Ausgabeverzeichnis für Bilder; Falls leer, werden die Pfade unterhalb verwendet",
-    "Output directory for txt2img images": "Ausgabeverzeichnis für txt2img Bilder",
-    "Output directory for img2img images": "Ausgabeverzeichnis für img2img Bilder",
-    "Output directory for images from extras tab": "Ausgabeverzeichnis für Extras-Tab Bilder",
-    "Output directory for grids; if empty, defaults to two directories below": "Ausgabeverzeichnis für Raster; Falls leer, werden die Pfade unterhalb verwendet",
-    "Output directory for txt2img grids": "Ausgabeverzeichnis für txt2img Raster",
-    "Output directory for img2img grids": "Ausgabeverzeichnis für img2img Raster",
-    "Directory for saving images using the Save button": "Ausgabeverzeichnis für Bilder, die mit dem 'Speichern'-Knopf gespeichert wurden",
-    "Saving to a directory": "Speichern in ein Verzeichnis",
-    "Save images to a subdirectory": "Bilder in ein Unterverzeichnis speichern",
-    "Save grids to a subdirectory": "Raster in ein Unterverzeichnis speichern",
-    "When using \"Save\" button, save images to a subdirectory": "Bilder bei der Benutzung des 'Speichern'-Knopfes in ein Unterverzeichnis speichern",
-    "Directory name pattern": "Muster für Verzeichnisnamen",
-    "Max prompt words for [prompt_words] pattern": "Maximale Wortanzahl für [prompt_words] Muster",
-    "Upscaling": "Upscaling",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "Kachelgröße für ESRGAN-Upscaler. 0 = keine Kacheln.",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "Kachelüberlappung in Pixeln für ESRGAN-Upscaler. Niedrige Werte = sichtbare Naht.",
-    "Tile size for all SwinIR.": "Kachelgröße für alle SwinIR.",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "Kachelüberlappung in Pixeln für SwinIR. Niedrige Werte = sichtbare Naht.",
-    "LDSR processing steps. Lower = faster": "LDSR-Verarbeitungsschritte. Niedriger = schneller",
-    "Upscaler for img2img": "Upscaler für img2img",
-    "Upscale latent space image when doing hires. fix": "Bild des Latent Space upscalen, wenn Highres. Fix benutzt wird",
-    "Face restoration": "Gesichtswiederhestellung",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormer Gewichtung (0 = maximale Wirkung, 1 = minimale Wirkung)",
-    "Move face restoration model from VRAM into RAM after processing": "Verschiebe Gesichtswiederhestellung-Modell nach der Verarbeitung vom VRAM in den RAM",
-    "System": "System",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "VRAM-Nutzungsabfragen pro Sekunde während der Generierung. Zum Deaktivieren auf 0 setzen.",
-    "Always print all generation info to standard output": "Immer alle Generationsinformationen in der Standardausgabe ausgeben",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Der Konsole einen zweiten Fortschrittsbalken hinzufügen, der den Fortschritt eines gesamten Auftrags anzeigt.",
-    "Training": "Training",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "VAE und CLIP während des Hypernetwork-Trainings in den RAM verschieben. Spart VRAM.",
-    "Filename word regex": "Filename word regex",
-    "Filename join string": "Filename join string",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Anzahl der Wiederholungen für ein einzelnes Eingabebild pro Epoche; wird nur für die Anzeige der Epochennummer verwendet",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Speichere eine csv-Datei, die den Verlust enthält, im Protokollverzeichnis alle N Schritte, 0 zum Deaktivieren",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "Checkpoints zum Zwischenspeichern im RAM",
-    "Hypernetwork strength": "Hypernetworkstärke",
-    "Inpainting conditioning mask strength": "Inpainting Stärke der Konditionierungsmaske",
-    "Apply color correction to img2img results to match original colors.": "Farbkorrektur auf die img2img-Ergebnisse anwenden, damit sie den Originalfarben entsprechen.",
-    "Save a copy of image before applying color correction to img2img results": "Vor dem Anwenden der Farbkorrektur eine Kopie des Bildes speichern",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "Mit img2img, die exakte Anzahl der Schritte ausführen, die vom Schieberegler angegeben sind (normalerweise weniger bei weniger Denoising).",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Aktivieren der Quantisierung in K-Samplern für schärfere und sauberere Ergebnisse. Dies kann bestehende Seeds verändern. Erfordert Neustart.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Hervorhebung: Verwenden Sie (Text), damit das Modell dem Text mehr Aufmerksamkeit schenkt, und [Text], damit es ihm weniger Aufmerksamkeit schenkt",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Verwenden der alten Implementierung von Hervorhebungen. Kann nützlich sein, um alte Seeds zu reproduzieren.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "K-Diffusions-Sampler erzeugen in einem Batch die gleichen Bilder, wie bei der Erstellung eines einzelnen Bildes",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "Erhöhung der Kohärenz durch Auffüllen ab dem letzten Komma innerhalb von n Token, wenn mehr als 75 Token verwendet werden",
-    "Filter NSFW content": "NSFW-Inhalte filtern",
-    "Stop At last layers of CLIP model": "Stoppe bei den letzten Schichten des CLIP-Modells",
-    "Interrogate Options": "Interrogate Optionen",
-    "Interrogate: keep models in VRAM": "Interrogate: Modelle im VRAM behalten",
-    "Interrogate: use artists from artists.csv": "Interrogate: Künstler aus 'artists.csv' nutzen",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interrogate: Die Rangfolge von Modell-Tags in den Ergebnissen einschließen (hat keine Auswirkung auf beschriftungsbasierte Interrogator).",
-    "Interrogate: num_beams for BLIP": "Interrogate: num_beams für BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Interrogate: minimale Beschreibungslänge (Künstler, etc.. ausgenommen)",
-    "Interrogate: maximum description length": "Interrogate: maximale Beschreibungslänge",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: maximale Anzahl an Zeilen in Textdatei (0 = Kein Limit)",
-    "Interrogate: deepbooru score threshold": "Interrogate: Deepbooru minimale Punkteanzahl",
-    "Interrogate: deepbooru sort alphabetically": "Interrogate: Sortiere Deepbooru alphabetisch",
-    "use spaces for tags in deepbooru": "Benutze Leerzeichen für Deepbooru-Tags",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "Escape-Klammern (\\) in Deepbooru (damit sie als normale Klammern und nicht zur Hervorhebung verwendet werden)",
-    "User interface": "Benutzeroberfläche",
-    "Show progressbar": "Fortschrittsleiste anzeigen",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Zeige eine Bildvorschau alle N Samplingschritte. Zum Deaktivieren auf 0 setzen.",
-    "Show previews of all images generated in a batch as a grid": "Zeige eine Vorschau aller erzeugten Bilder in einem Batch als Raster",
-    "Show grid in results for web": "Zeige Raster in der Web-UI Vorschau",
-    "Do not show any images in results for web": "Keine Bilder in der Web-UI Vorschau zeigen",
-    "Add model hash to generation information": "Hash des Modells zu den Generationsinformationen hinzufügen",
-    "Add model name to generation information": "Name des Modells zu den Generationsinformationen hinzufügen",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "Beim Einlesen von Generierungsparametern aus Text in die Benutzeroberfläche (aus PNG-Info oder eingefügtem Text) wird das ausgewählte Modell/Checkpoint nicht geändert.",
-    "Send seed when sending prompt or image to other interface": "Den Seed, beim Senden des Bildes/Prompt zu einem anderen Tab, mitsenden",
-    "Font for image grids that have text": "Schriftart für Bildraster mit Text",
-    "Enable full page image viewer": "Ganzseitenbildbetrachter aktivieren",
-    "Show images zoomed in by default in full page image viewer": "Standardmäßig Bilder im Ganzseitenbildbetrachter vergrößert anzeigen",
-    "Show generation progress in window title.": "Generationsfortschritt im Fenstertitel anzeigen.",
-    "Quicksettings list": "Schnellzugriffsleiste",
-    "Localization (requires restart)": "Lokalisierung (Erfordert Neustart)",
-    "Sampler parameters": "Samplerparameter",
-    "Hide samplers in user interface (requires restart)": "Sampler in der Benutzeroberfläche verstecken (Erfordert Neustart)",
-    "eta (noise multiplier) for DDIM": "Eta (noise Multiplikator) für DDIM",
-    "eta (noise multiplier) for ancestral samplers": "Eta (noise Multiplikator) für Ancestral Sampler",
-    "img2img DDIM discretize": "img2img DDIM diskretisieren",
-    "uniform": "uniform",
-    "quad": "quad",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma noise",
-    "Eta noise seed delta": "Eta noise seed delta",
-    "Request browser notifications": "Browserbenachrichtigungen anfordern",
-    "Download localization template": "Vorlage für Lokalisierung herunterladen",
-    "Reload custom script bodies (No ui updates, No restart)": "Benutzerdefinierte Skripte neu laden (keine Aktualisierung der Benutzeroberfläche, kein Neustart)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Gradio neu starten und Komponenten aktualisieren (nur Custom Scripts, ui.py, js und css)",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "Prompt (zum Erzeugen Strg+Eingabe oder Alt+Eingabe drücken)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Negative Prompt (zum Erzeugen Strg+Eingabe oder Alt+Eingabe drücken)",
-    "Add a random artist to the prompt.": "Zufälligen Künstler der Prompt hinzufügen.",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Lesen der Generationsparameter aus der Prompt oder der letzten Generation (wenn Prompt leer ist) in die Benutzeroberfläche.",
-    "Save style": "Stil speichern",
-    "Apply selected styles to current prompt": "Momentan ausgewählte Stile auf die Prompt anwenden",
-    "Stop processing current image and continue processing.": "Verarbeitung des momentanen Bildes abbrechen und Verarbeitung fortsetzen.",
-    "Stop processing images and return any results accumulated so far.": "Verarbeitung abbrechen und alle bisherigen Ergebnisse ausgeben.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Stil, der angwendet werden soll. Stile haben sowohl positive als auch negative Promptanteile und werden auf beide angewandt.",
-    "Do not do anything special": "Nichts besonderes machen",
-    "Which algorithm to use to produce the image": "Der zu benutzende Algorithmus für die Bildgeneration",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - sehr kreativ, kann sehr unterschiedliche Bilder in Abhängigkeit von der Schrittanzahl bekommen. Werte höher als 30-40 helfen nicht.",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Modelle - am besten für inpainting",
-    "Produce an image that can be tiled.": "Bild erzeugen, dass gekachelt werden kann.",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Verwendung eines zweistufigen Prozesses, um ein Bild mit geringerer Auflösung zu erstellen, hochzuskalieren und dann die Details zu verbessern, ohne die Komposition zu verändern.",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Bestimmt, wie wenig Bezug der Algorithmus zu dem Inhalt des Bildes haben soll. Bei 0 ändert sich nichts, und bei 1 besitzt das Bild keinen Bezug. Bei Werten unter 1,0 erfolgt die Verarbeitung in weniger Schritten, als der Schieberegler angibt.",
-    "How many batches of images to create": "Wie viele Sätze von Bildern erstellt werden sollen",
-    "How many image to create in a single batch": "Wie viele Bilder in einem Batch erstellt werden sollen",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - wie stark das Bild der Prompt entsprechen soll - niedrigere Werte führen zu kreativeren Ergebnissen",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Ein Wert, der die Ausgabe des Zufallszahlengenerators bestimmt: Wenn ein Bild mit denselben Parametern und demselben Seed wie ein anderes Bild erstellt wird, erhält man dasselbe Ergebnis.",
-    "Set seed to -1, which will cause a new random number to be used every time": "Seed auf -1 setzen, so dass jedes Mal eine neue Zufallszahl verwendet wird",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Wiederverwendung des Seeds der letzten Generation, meist nützlich, wenn er zufällig gewählt wurde",
-    "Seed of a different picture to be mixed into the generation.": "Seed eines anderen Bildes, der bei der Erzeugung reingemischt wird.",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Wie stark die Veränderung sein soll. Bei 0 gibt es keinen Effekt. Bei 1 erhält man das vollständige Bild mit dem Variationsseed (außer bei Ancestral Samplern, wie Euler A, wo man nur etwas erhält).",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Versuche ein Bild zu erzeugen, das dem ähnelt, das mit dem Seed bei der angegebenen Auflösung erzeugt worden wäre.",
-    "Separate values for X axis using commas.": "Trenne die Werte für die X-Achse durch Kommas.",
-    "Separate values for Y axis using commas.": "Trenne die Werte für die Y-Achse durch Kommas.",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Bild in ein Verzeichnis (Standard - log/images) und Generationsparameter in eine csv-Datei schreiben.",
-    "Open images output directory": "Ausgabeverzeichnis öffnen",
-    "How much to blur the mask before processing, in pixels.": "Wie stark die Maske vor der Verarbeitung weichgezeichnet werden soll, in Pixeln.",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Was in den maskierten Bereich vor der Verarbeitung mit Stable Diffusion soll.",
-    "fill it with colors of the image": "Füllen mit den Farben des Bildes",
-    "keep whatever was there originally": "Originalen Inhalt behalten",
-    "fill it with latent space noise": "Füllen mit latent space noise",
-    "fill it with latent space zeroes": "Füllen mit latent space Nullen",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "Hochskalieren des maskierten Bereichs auf die Zielauflösung, Inpainting, Zurückskalieren und Einfügen in das Originalbild.",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Die Größe des Bildes auf die gewünschte Auflösung ändern. Wenn Höhe und Breite nicht übereinstimmen, erhält man ein falsches Seitenverhältnis.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Die Größe des Bildes so ändern, dass die gesamte Zielauflösung mit dem Bild ausgefüllt wird. Herausragende Teile werden abgeschnitten.",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Die Größe des Bildes so ändern, dass das gesamte Bild enthalten ist. Lücken werden mit Farben des Bildes ausgefüllt.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Wie oft die Verarbeitung eines Bildes wiederholt und als Eingabe für die nächste Iteration verwendet werden soll",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "Im Loopback-Modus wird die Denoisingstärke in jeder Schleife mit diesem Wert multipliziert. <1 bedeutet abnehmende Vielfalt, so dass die Sequenz zu einem festen Bild konvergiert. >1 bedeutet zunehmende Vielfalt, so dass die Sequenz immer chaotischer wird.",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "Wie viel Pixel sich beim SD-Upscale zwischen den Kacheln überlappen. Die Kacheln überlappen sich so, dass beim Zusammenfügen zu einem Bild keine deutlich sichtbare Naht entsteht.",
-    "A directory on the same machine where the server is running.": "Ein Verzeichnis auf demselben Rechner, auf dem der Server läuft.",
-    "Leave blank to save images to the default path.": "Leer lassen, um Bilder im Standardpfad zu speichern.",
-    "Result = A * (1 - M) + B * M": "Ergebnis = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Ergebnis = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "Erste und letzte Ziffer müssen 1 sein. Bspl:'1, 2, 1'",
-    "Path to directory with input images": "Pfad zum Verzeichnis mit den Eingabebildern",
-    "Path to directory where to write outputs": "Pfad zum Verzeichnis, wo die Ausgaben gespeichert werden",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Verwende die folgenden Tags, um festzulegen, wie die Dateinamen für Bilder ausgewählt werden: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leer lassen, um Standardwerte zu verwenden.",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Wenn diese Option aktiviert ist, wird den erstellten Bildern kein Wasserzeichen hinzugefügt. Achtung: Wenn Sie kein Wasserzeichen hinzufügen, verhalten Sie sich möglicherweise unethisch.",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Verwenden Sie die folgenden Tags, um festzulegen, wie Unterverzeichnisse für Bilder und Raster ausgewählt werden: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leer lassen, um Standardwerte zu verwenden.",
-    "Restore low quality faces using GFPGAN neural network": "Wiederherstellung von Gesichtern schlechter Qualität mit dem neuralen Netzwerk GFPGAN",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Dieser reguläre Ausdruck wird verwendet, um Wörter aus dem Dateinamen zu extrahieren, die dann mit der unten stehenden Option zu einem Beschriftungstext für das Training zusammengefügt werden. Leer lassen, um den Text des Dateinamens so zu belassen, wie er ist.",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Diese Zeichenfolge wird verwendet, um getrennte Wörter in einer einzigen Zeile zu verbinden, wenn die obige Option aktiviert ist.",
-    "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "Gilt nur für Inpainting-Modelle. Legt fest, wie stark das Originalbild für Inpainting und img2img maskiert werden soll. 1.0 bedeutet vollständig maskiert, was das Standardverhalten ist. 0.0 bedeutet eine vollständig unmaskierte Konditionierung. Niedrigere Werte tragen dazu bei, die Gesamtkomposition des Bildes zu erhalten, sind aber bei großen Änderungen problematisch.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Liste von Einstellungsnamen, getrennt durch Kommas, für Einstellungen, die in der Schnellzugriffsleiste oben erscheinen sollen, anstatt in dem üblichen Einstellungs-Tab. Siehe modules/shared.py für Einstellungsnamen. Erfordert einen Neustart zur Anwendung.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Wenn dieser Wert ungleich Null ist, wird er zum Seed addiert und zur Initialisierung des RNG für Noise bei der Verwendung von Samplern mit Eta verwendet. Dies kann verwendet werden, um noch mehr Variationen von Bildern zu erzeugen, oder um Bilder von anderer Software zu erzeugen, wenn Sie wissen, was Sie tun."
-}

localizations/es_ES.json

@@ -1,692 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "Cargando...",
-    "view": "Mostrar ",
-    "api": "API",
-    "•": " • ",
-    "Construido con Gradio": "Construido con Gradio",
-    "Stable Diffusion checkpoint": "Stable Diffusion checkpoint",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "Extras",
-    "PNG Info": "Info PNG",
-    "Checkpoint Merger": "Fusionar Checkpoints",
-    "Train": "Entrenar",
-    "Deforum": "Deforum",
-    "Image Browser": "Navegador de Imágenes",
-    "Settings": "Ajustes",
-    "Extensions": "Extensiones",
-    "Prompt": "Prompt",
-    "Negative prompt": "Prompt negativo",
-    "Run": "Ejecutar",
-    "Skip": "Saltar",
-    "Interrupt": "Interrumpir",
-    "Generate": "Generar",
-    "Style 1": "Estilo 1",
-    "Style 2": "Estilo 2",
-    "Label": "Etiqueta",
-    "File": "Archivo",
-    "Coloque el archivo aquí": "Suelta el archivo aquí",
-    "-": "-",
-    "o": "o",
-    "Haga click para cargar": "Haz click para cargar",
-    "Image": "Imagen",
-    "Check progress": "Comprobar progreso",
-    "Check progress (first)": "Comprobar progreso (inicial)",
-    "Sampling Steps": "Sampling Steps",
-    "Sampling method": "Método de Sampling",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Ancho",
-    "Height": "Alto",
-    "Restore faces": "Restaurar rostros",
-    "Tiling": "Mosaico",
-    "Highres. fix": "Highres. fix",
-    "Firstpass width": "Ancho original",
-    "Firstpass height": "Alto original",
-    "Denoising strength": "Denoising strength",
-    "Batch count": "Cantidad del Batch",
-    "Batch size": "Tamaño del Batch",
-    "CFG Scale": "Escala CFG",
-    "Seed": "Seed",
-    "Extra": "Extra",
-    "Variation seed": "Seed de variación",
-    "Variation strength": "Fuerza de variación",
-    "Resize seed from width": "Redimensionar seed del ancho",
-    "Resize seed from height": "Redimensionar seed del alto",
-    "Script": "Script",
-    "None": "Ninguno",
-    "Prompt matrix": "Prompt en matriz",
-    "Prompts from file or textbox": "Prompts desde archivo o campo de texto",
-    "X/Y plot": "Tabla X/Y",
-    "Put variable parts at start of prompt": "Poner partes variables al inicio del prompt",
-    "Iterate seed every line": "Repetir seed en cada línea",
-    "Use same random seed for all lines": "Utiliza el mismo seed aleatorio para todas las líneas",
-    "List of prompt inputs": "Listado de prompts",
-    "Upload prompt inputs": "Cargar archivo de prompts",
-    "X type": "X",
-    "Nothing": "Nada",
-    "Var. seed": "Var. seed",
-    "Var. strength": "Var. fuerza",
-    "Steps": "Steps",
-    "Prompt S/R": "Prompt S/R",
-    "Prompt order": "Prompt order",
-    "Sampler": "Sampler",
-    "Checkpoint name": "Nombre Checkpoint",
-    "Hypernetwork": "Hypernetwork",
-    "Hypernet str.": "Hypernet str.",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma min",
-    "Sigma max": "Sigma max",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Eta",
-    "Clip skip": "Clip skip",
-    "Denoising": "Denoising",
-    "Cond. Image Mask Weight": "Cond. Image Mask Weight",
-    "X values": "Valores X",
-    "Y type": "Y",
-    "Y values": "Valores Y",
-    "Draw legend": "Agregar leyenda",
-    "Include Separate Images": "Incluir Imágenes Separadas",
-    "Keep -1 for seeds": "Mantener -1 para seeds",
-    "Save": "Guardar",
-    "Send to img2img": "Enviar a img2img",
-    "Send to inpaint": "Enviar a inpaint",
-    "Send to extras": "Enviar a extras",
-    "Make Zip when Save?": "Crear Zip al Guardar?",
-    "Textbox": "Campo de texto",
-    "Interrogate\nCLIP": "Interrogar\nCLIP",
-    "Inpaint": "Inpaint",
-    "Batch img2img": "Batch img2img",
-    "Image for img2img": "Imagen para img2img",
-    "Coloque la imagen aquí": "Suelta la imagen aquí",
-    "Image for inpainting with mask": "Imagen para inpainting con máscara",
-    "Mask": "Máscara",
-    "Mask blur": "Difuminar máscara",
-    "Mask mode": "Modo máscara",
-    "Draw mask": "Dibujar máscara",
-    "Upload mask": "Cargar máscara",
-    "Masking mode": "Modo de enmascarado",
-    "Inpaint masked": "Inpaint con enmascarado",
-    "Inpaint not masked": "Inpaint sin enmascarado",
-    "Masked content": "Contenido enmascarado",
-    "fill": "rellenar",
-    "original": "original",
-    "latent noise": "latent noise",
-    "latent nothing": "latent nothing",
-    "Inpaint at full resolution": "Inpaint a resolución completa",
-    "Inpaint at full resolution padding, pixels": "Inpaint a resolución completa con relleno, en pixeles",
-    "Process images in a directory on the same machine where the server is running.": "Procesa imágenes en un directorio en la misma máquina donde se ejecuta el servidor.",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Usa un directorio de salida vacío para guardar imágenes normalmente en lugar de escribir en el directorio de salida.",
-    "Input directory": "Directorio de entrada",
-    "Output directory": "Directorio de salida",
-    "Resize mode": "Modo de cambio de tamaño",
-    "Just resize": "Solo redimensionar",
-    "Crop and resize": "Recortar y redimensionar",
-    "Resize and fill": "Redimensionar y rellenar",
-    "img2img alternative test": "img2img alternative test",
-    "Loopback": "Loopback",
-    "Outpainting mk2": "Outpainting mk2",
-    "Poor man's outpainting": "Poor man's outpainting",
-    "SD upscale": "SD upscale",
-    "Deforum-webui (use tab extension instead!)": "Deforum-webui (utiliza la extensión en su lugar!)",
-    "should be 2 or lower.": "debe ser 2 o menos.",
-    "Override `Sampling method` to Euler?(this method is built for it)": "Invalidar `Sampling method` a Euler? (este método está diseñado para ello)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "Invalidar `prompt` al mismo valor `prompt original`? (y `prompt negativo`)",
-    "Original prompt": "Prompt original",
-    "Original negative prompt": "Promp negativo original",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "Invalidar `Sampling Steps` al mismo valor de `Decode steps`?",
-    "Decode steps": "Decode steps",
-    "Override `Denoising strength` to 1?": "Invalidar `Denoising strength` a 1?",
-    "Decode CFG scale": "Decodificar escala CFG",
-    "Randomness": "Aleatoriedad",
-    "Sigma adjustment for finding noise for image": "Ajuste Sigma para encontrar ruido para la imagen.",
-    "Loops": "Loops",
-    "Denoising strength change factor": "Denoising strength change factor",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Ajustes recomendados: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8",
-    "Pixels to expand": "Píxeles para expandir",
-    "Outpainting direction": "Dirección Outpainting",
-    "left": "izquierda",
-    "right": "derecha",
-    "up": "arriba",
-    "down": "abajo",
-    "Fall-off exponent (lower=higher detail)": "Fall-off exponent (inferior=mayor detalle)",
-    "Color variation": "Variación de color",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Mejorará la imagen al doble de las dimensiones; usa los controles deslizantes de ancho y alto para establecer el tamaño del mosaico",
-    "Tile overlap": "Solapar mosaicos",
-    "Upscaler": "Upscaler",
-    "Lanczos": "Lanczos",
-    "Nearest": "Nearest",
-    "LDSR": "LDSR",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR 4x": "SwinIR 4x",
-    "Deforum v0.5-webui-beta": "Deforum v0.5-webui-beta",
-    "This script is deprecated. Please use the full Deforum extension instead.": "Este script está obsoleto. Utiliza la extensión completa de Deforum en su lugar.",
-    "Update instructions:": "Instrucciones para actualizar:",
-    "github.com/deforum-art/deforum-for-automatic1111-webui/blob/automatic1111-webui/README.md": "github.com/deforum-art/deforum-for-automatic1111-webui/blob/automatic1111-webui/README.md",
-    "discord.gg/deforum": "discord.gg/deforum",
-    "Single Image": "Imagen Única",
-    "Batch Process": "Batch Process",
-    "Batch from Directory": "Batch desde Directorio",
-    "Source": "Origen",
-    "Show result images": "Mostrar imágenes generadas",
-    "Scale by": "Escalar por",
-    "Scale to": "Escalar a",
-    "Resize": "Redimensionar",
-    "Crop to fit": "Recortar para ajustar",
-    "Upscaler 2 visibility": "Visibilidad Upscaler 2",
-    "GFPGAN visibility": "Visibilidad GFPGAN",
-    "CodeFormer visibility": "Visibilidad CodeFormer",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "Influencia CodeFormer (0 = efecto máximo, 1 = efecto mínimo)",
-    "Upscale Before Restoring Faces": "Escalar antes de Restaurar Rostros",
-    "Send to txt2img": "Enviar a txt2img",
-    "A merger of the two checkpoints will be generated in your": "Se generará una fusión de los dos checkpoints en tu ",
-    "checkpoint": "directorio ",
-    "directory.": "de modelos.",
-    "Primary model (A)": "Modelo primario (A)",
-    "Secondary model (B)": "Modelo secundario (B)",
-    "Tertiary model (C)": "Modelo terciario (C)",
-    "Custom Name (Optional)": "Nombre personalizado (Opcional)",
-    "Multiplier (M) - set to 0 to get model A": "Multiplier (M) - establecer en 0 para obtener el modelo A",
-    "Interpolation Method": "Método de interpolación",
-    "Weighted sum": "Weighted sum",
-    "Add difference": "Add difference",
-    "Save as float16": "Guardar como float16",
-    "See": "Ver ",
-    "wiki": "wiki ",
-    "for detailed explanation.": "para una explicación detallada.",
-    "Create embedding": "Crear embedding",
-    "Create hypernetwork": "Crear hypernetwork",
-    "Preprocess images": "Preprocesar imágenes",
-    "Name": "Nombre",
-    "Initialization text": "Texto de inicialización",
-    "Number of vectors per token": "Número de vectores por token",
-    "Overwrite Old Embedding": "Sobrescribir Embedding Anterior",
-    "Modules": "Módulos",
-    "Enter hypernetwork layer structure": "Ingresa la estructura de capa del hypernetwork",
-    "Select activation function of hypernetwork": "Selecciona la función de activación del hypernetwork",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "tanh": "tanh",
-    "sigmoid": "sigmoid",
-    "celu": "celu",
-    "gelu": "gelu",
-    "glu": "glu",
-    "hardshrink": "hardshrink",
-    "hardsigmoid": "hardsigmoid",
-    "hardtanh": "hardtanh",
-    "logsigmoid": "logsigmoid",
-    "logsoftmax": "logsoftmax",
-    "mish": "mish",
-    "prelu": "prelu",
-    "rrelu": "rrelu",
-    "relu6": "relu6",
-    "selu": "selu",
-    "silu": "silu",
-    "softmax": "softmax",
-    "softmax2d": "softmax2d",
-    "softmin": "softmin",
-    "softplus": "softplus",
-    "softshrink": "softshrink",
-    "softsign": "softsign",
-    "tanhshrink": "tanhshrink",
-    "threshold": "threshold",
-    "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "Seleccionar inicialización de modelos de capa. relu-like - Kaiming, sigmoid-like - Xavier es el recomendado",
-    "Normal": "Normal",
-    "KaimingUniform": "KaimingUniform",
-    "KaimingNormal": "KaimingNormal",
-    "XavierUniform": "XavierUniform",
-    "XavierNormal": "XavierNormal",
-    "Add layer normalization": "Agregar normalización de capa",
-    "Use dropout": "Usar dropout",
-    "Overwrite Old Hypernetwork": "Sobrescribir Hypernetwork Anterior",
-    "Source directory": "Directorio de origen",
-    "Destination directory": "Directorio de salida",
-    "Existing Caption txt Action": "Existing Caption txt Action",
-    "ignore": "ignorar",
-    "copy": "copiar",
-    "prepend": "anteponer",
-    "append": "añadir",
-    "Create flipped copies": "Crear copias volteadas",
-    "Split oversized images": "Dividir imágenes muy grandes",
-    "Auto focal point crop": "Recorte de punto focal automático",
-    "Use BLIP for caption": "Usar BLIP para leyenda",
-    "Use deepbooru for caption": "Usar deepbooru para leyenda",
-    "Split image threshold": "Umbral en imagen dividida",
-    "Split image overlap ratio": "Relación de superposición en imagen dividida",
-    "Focal point face weight": "Peso de la cara del punto focal",
-    "Focal point entropy weight": "Focal point entropy weight",
-    "Focal point edges weight": "Focal point edges weight",
-    "Create debug image": "Crear imagen de depuración",
-    "Preprocess": "Preproceso",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "Entrenar un embedding o Hypernetwork; debes especificar un directorio con un conjunto de imágenes con una proporción de 1:1 ",
-    "[wiki]": "[wiki]",
-    "Embedding": "Embedding",
-    "Embedding Learning rate": "Embedding Learning rate",
-    "Hypernetwork Learning rate": "Hypernetwork Learning rate",
-    "Dataset directory": "Directorio dataset",
-    "Log directory": "Directorio log",
-    "Prompt template file": "Prompt archivos plantilla",
-    "Max steps": "Max steps",
-    "Save an image to log directory every N steps, 0 to disable": "Guarda una imagen en el directorio log cada N pasos, 0 para deshabilitar",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Guarda una copia de embedding en el directorio log cada N pasos, 0 para deshabilitar",
-    "Save images with embedding in PNG chunks": "Guarda imágenes con embedding en fragmentos PNG",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Leer parámetros (prompt, etc...) desde la pestaña txt2img al hacer vistas previas",
-    "Train Hypernetwork": "Entrenar Hypernetwork",
-    "Train Embedding": "Entrenar Embedding",
-    "Info and links": "Información y enlaces",
-    "▼": "▼",
-    "Made by deforum.github.io, port for AUTOMATIC1111's webui maintained by kabachuha": "Desarrolado por deforum.github.io, port para AUTOMATIC1111's webui mantenido por kabachuha",
-    "Original Deforum Github repo  github.com/deforum/stable-diffusion": "Deforum Github repo  github.com/deforum/stable-diffusion",
-    "This fork for auto1111's webui github.com/deforum-art/deforum-for-automatic1111-webui": "Fork para auto1111's webui github.com/deforum-art/deforum-for-automatic1111-webui",
-    "Join the official Deforum Discord discord.gg/deforum to share your creations and suggestions": "Únete al Discord oficial Deforum discord.gg/deforum para compartir tus creaciones y sugerencias",
-    "User guide for v0.5 docs.google.com/document/d/1pEobUknMFMkn8F5TMsv8qRzamXX_75BShMMXV8IFslI/edit": "Guía de usuario v0.5 docs.google.com/document/d/1pEobUknMFMkn8F5TMsv8qRzamXX_75BShMMXV8IFslI/edit",
-    "Math keyframing explanation docs.google.com/document/d/1pfW1PwbDIuW0cv-dnuyYj1UzPqe23BlSLTJsqazffXM/edit?usp=sharing": "Explicación de fotogramas matemáticos docs.google.com/document/d/1pfW1PwbDIuW0cv-dnuyYj1UzPqe23BlSLTJsqazffXM/edit?usp=sharing",
-    "Keyframes": "Keyframes",
-    "Prompts": "Prompts",
-    "Init": "Init",
-    "Video output": "Salida de vídeo",
-    "Run settings": "Ejecutar ajustes",
-    "Import settings from file": "Importar ajustes desde archivo",
-    "Override settings": "Invalidar ajustes",
-    "Custom settings file": "Archivo de ajustes personalizados",
-    "Sampling settings": "Ajustes de sampling",
-    "override_these_with_webui": "override_these_with_webui",
-    "W": "Ancho",
-    "H": "Alto",
-    "seed": "seed",
-    "sampler": "sampler",
-    "Enable extras": "Activar extras",
-    "subseed": "subseed",
-    "subseed_strength": "subseed_strength",
-    "steps": "steps",
-    "ddim_eta": "ddim_eta",
-    "n_batch": "n_batch",
-    "make_grid": "make_grid",
-    "grid_rows": "grid_rows",
-    "save_settings": "save_settings",
-    "save_samples": "save_samples",
-    "display_samples": "display_samples",
-    "save_sample_per_step": "save_sample_per_step",
-    "show_sample_per_step": "show_sample_per_step",
-    "Batch settings": "Ajustes de batch",
-    "batch_name": "batch_name",
-    "filename_format": "filename_format",
-    "seed_behavior": "seed_behavior",
-    "iter": "iter",
-    "fixed": "fixed",
-    "random": "random",
-    "schedule": "schedule",
-    "Animation settings": "Ajustes de animación",
-    "animation_mode": "animation_mode",
-    "2D": "2D",
-    "3D": "3D",
-    "Video Input": "Entrada de Video",
-    "max_frames": "max_frames",
-    "border": "border",
-    "replicate": "replicate",
-    "wrap": "wrap",
-    "Motion parameters:": "Parámetros de movimiento:",
-    "2D and 3D settings": "Ajustes 2D y 3D",
-    "angle": "angle",
-    "zoom": "zoom",
-    "translation_x": "translation_x",
-    "translation_y": "translation_y",
-    "3D settings": "Ajustes 3D",
-    "translation_z": "translation_z",
-    "rotation_3d_x": "rotation_3d_x",
-    "rotation_3d_y": "rotation_3d_y",
-    "rotation_3d_z": "rotation_3d_z",
-    "Prespective flip — Low VRAM pseudo-3D mode:": "Prespective flip — Modo Low VRAM pseudo-3D:",
-    "flip_2d_perspective": "flip_2d_perspective",
-    "perspective_flip_theta": "perspective_flip_theta",
-    "perspective_flip_phi": "perspective_flip_phi",
-    "perspective_flip_gamma": "perspective_flip_gamma",
-    "perspective_flip_fv": "perspective_flip_fv",
-    "Generation settings:": "Ajustes de generación:",
-    "noise_schedule": "noise_schedule",
-    "strength_schedule": "strength_schedule",
-    "contrast_schedule": "contrast_schedule",
-    "cfg_scale_schedule": "cfg_scale_schedule",
-    "3D Fov settings:": "Ajustes 3D Fov:",
-    "fov_schedule": "fov_schedule",
-    "near_schedule": "near_schedule",
-    "far_schedule": "far_schedule",
-    "To enable seed schedule select seed behavior — 'schedule'": "Para habilitar el seed schedule, selecciona el comportamiento del seed — 'schedule'",
-    "seed_schedule": "seed_schedule",
-    "Coherence:": "Coherencia:",
-    "color_coherence": "color_coherence",
-    "Match Frame 0 HSV": "Match Frame 0 HSV",
-    "Match Frame 0 LAB": "Match Frame 0 LAB",
-    "Match Frame 0 RGB": "Match Frame 0 RGB",
-    "diffusion_cadence": "diffusion_cadence",
-    "3D Depth Warping:": "3D Depth Warping:",
-    "use_depth_warping": "use_depth_warping",
-    "midas_weight": "midas_weight",
-    "near_plane": "near_plane",
-    "far_plane": "far_plane",
-    "fov": "fov",
-    "padding_mode": "padding_mode",
-    "reflection": "reflection",
-    "zeros": "zeros",
-    "sampling_mode": "sampling_mode",
-    "bicubic": "bicubic",
-    "bilinear": "bilinear",
-    "nearest": "nearest",
-    "save_depth_maps": "save_depth_maps",
-    "`animation_mode: None` batches on list of *prompts*. (Batch mode disabled atm, only animation_prompts are working)": "`animation_mode: None` batches en lista de *prompts*. (Modo batch deshabilitado por el momento, solamente animation_prompts esta funcionando)",
-    "*Important change from vanilla Deforum!*": "*Cambios importantes en Deforum!*",
-    "This script uses the built-in webui weighting settings.": "Este script utiliza la configuración de pesos integrados.",
-    "So if you want to use math functions as prompt weights,": "Entonces, si deseas usar funciones matemáticas con pesos en los prompts,",
-    "keep the values above zero in both parts": "manten los valores por encima de cero en ambas partes",
-    "Negative prompt part can be specified with --neg": "La parte de prompt negativo se puede especificar utilizando --neg",
-    "batch_prompts (disabled atm)": "batch_prompts (deshabilitado por el momento)",
-    "animation_prompts": "animation_prompts",
-    "Init settings": "Ajustes Init",
-    "use_init": "use_init",
-    "from_img2img_instead_of_link": "from_img2img_instead_of_link",
-    "strength_0_no_init": "strength_0_no_init",
-    "strength": "strength",
-    "init_image": "init_image",
-    "use_mask": "use_mask",
-    "use_alpha_as_mask": "use_alpha_as_mask",
-    "invert_mask": "invert_mask",
-    "overlay_mask": "overlay_mask",
-    "mask_file": "mask_file",
-    "mask_brightness_adjust": "mask_brightness_adjust",
-    "mask_overlay_blur": "mask_overlay_blur",
-    "Video Input:": "Entrada de Video:",
-    "video_init_path": "video_init_path",
-    "extract_nth_frame": "extract_nth_frame",
-    "overwrite_extracted_frames": "overwrite_extracted_frames",
-    "use_mask_video": "use_mask_video",
-    "video_mask_path": "video_mask_path",
-    "Interpolation (turned off atm)": "Interpolación (apagado por el momento)",
-    "interpolate_key_frames": "interpolate_key_frames",
-    "interpolate_x_frames": "interpolate_x_frames",
-    "Resume animation:": "Reanudar animación:",
-    "resume_from_timestring": "resume_from_timestring",
-    "resume_timestring": "resume_timestring",
-    "Video output settings": "Ajustes video de salida",
-    "skip_video_for_run_all": "skip_video_for_run_all",
-    "fps": "fps",
-    "output_format": "output_format",
-    "PIL gif": "PIL gif",
-    "FFMPEG mp4": "FFMPEG mp4",
-    "ffmpeg_location": "ffmpeg_location",
-    "add_soundtrack": "add_soundtrack",
-    "soundtrack_path": "soundtrack_path",
-    "use_manual_settings": "use_manual_settings",
-    "render_steps": "render_steps",
-    "max_video_frames": "max_video_frames",
-    "path_name_modifier": "path_name_modifier",
-    "x0_pred": "x0_pred",
-    "x": "x",
-    "image_path": "image_path",
-    "mp4_path": "mp4_path",
-    "Click here after the generation to show the video": "Haz click aquí después de la generación para mostrar el video",
-    "NOTE: If the 'Generate' button doesn't work, go in Settings and click 'Restart Gradio and Refresh...'.": "NOTA: Si el botón 'Generar' no funciona, ve a los Ajustes y presiona 'Reinciar Gradio y Refrescar...'.",
-    "Save Settings": "Guardar Ajustes",
-    "Load Settings": "Cargar Ajustes",
-    "Path relative to the webui folder": "Ruta relativa al folder principal",
-    "Save Video Settings": "Guardar Ajustes de Video",
-    "Load Video Settings": "Cargar Ajustes de Video",
-    "Favorites": "Favoritos",
-    "Others": "Otros",
-    "Images directory": "Directorio de Imágenes",
-    "Dropdown": "Menú desplegable",
-    "First Page": "Primera Página",
-    "Prev Page": "Página Anterior",
-    "Page Index": "Índice de Página",
-    "Next Page": "Página Siguiente",
-    "End Page": "Última Página",
-    "delete next": "eliminar siguiente",
-    "Delete": "Eliminar",
-    "sort by": "ordenar por",
-    "path name": "nombre de ruta",
-    "date": "fecha",
-    "keyword": "palabra clave",
-    "Generate Info": "Generar Info",
-    "File Name": "Nombre de Archivo",
-    "Move to favorites": "Mover a favoritos",
-    "Renew Page": "Recargar Página",
-    "Number": "Número",
-    "set_index": "set_index",
-    "load_switch": "load_switch",
-    "turn_page_switch": "turn_page_switch",
-    "Checkbox": "Checkbox",
-    "Apply settings": "Aplicar ajustes",
-    "Saving images/grids": "Guardar imágenes/grids",
-    "Always save all generated images": "Siempre guardar imágenes generadas",
-    "File format for images": "Formato de archivo para imágenes",
-    "Images filename pattern": "Patrón en nombre archivo imágenes",
-    "Add number to filename when saving": "Agregar número al nombre de archivo al guardar",
-    "Always save all generated image grids": "Siempre guardar grids de imágenes generadas",
-    "File format for grids": "Formato de archivo para grids",
-    "Add extended info (seed, prompt) to filename when saving grid": "Agregar información extendida (seed, prompt) al nombre del archivo al guardar grid",
-    "Do not save grids consisting of one picture": "No guardar grids que consisten en una imagen",
-    "Prevent empty spots in grid (when set to autodetect)": "Evitar espacios vacíos en grids (cuando se establece detección automática)",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Recuento de filas de grids; usar -1 para la detección automática y 0 para que sea igual al batch size",
-    "Save text information about generation parameters as chunks to png files": "Guardar información de texto sobre parámetros de generación como fragmentos en archivos png",
-    "Create a text file next to every image with generation parameters.": "Crear un archivo de texto junto a cada imagen con parámetros de generación.",
-    "Save a copy of image before doing face restoration.": "Guardar una copia de la imagen antes de restaurar rostro.",
-    "Save a copy of image before applying highres fix.": "Guardar una copia de la imagen antes de aplicar highres fix.",
-    "Save a copy of image before applying color correction to img2img results": "Guarda una copia de la imagen antes de aplicar la corrección de color a los resultados de img2img",
-    "Quality for saved jpeg images": "Calidad para imágenes jpeg guardadas",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "Si la imagen PNG es más grande de 4 MB o cualquier dimensión es más grande que 4000, reduce la escala y guarda la copia como JPG",
-    "Use original name for output filename during batch process in extras tab": "Use el nombre original para el nombre del archivo de salida durante el batch process en la pestaña extras",
-    "When using 'Save' button, only save a single selected image": "Al usar el botón 'Guardar', solo guarda una sola imagen seleccionada",
-    "Do not add watermark to images": "No agregar watermark a las imágenes",
-    "Paths for saving": "Directorios para guardar",
-    "Output directory for images; if empty, defaults to three directories below": "Directorio de imágenes; si está vacío, se utilizan los siguientes 3 directorios",
-    "Output directory for txt2img images": "Directorio para guardar imágenes de txt2img",
-    "Output directory for img2img images": "Directorio para guardar imágenes de img2img",
-    "Output directory for images from extras tab": "Directorio para guardar imágenes de extras",
-    "Output directory for grids; if empty, defaults to two directories below": "Directorio de grids; si está vacío, se utilizan los siguientes 2 directorios",
-    "Output directory for txt2img grids": "Directorio para guardar txt2img grids",
-    "Output directory for img2img grids": "Directorio para guardar img2img grids",
-    "Directory for saving images using the Save button": "Directorio para guardar imágenes usando el botón Guardar",
-    "Saving to a directory": "Guardando a un directorio",
-    "Save images to a subdirectory": "Guardar imágenes a un subdirectorio",
-    "Save grids to a subdirectory": "Guardar grids a un subdirectorio",
-    "When using \"Save\" button, save images to a subdirectory": "Al usar el botón \"Guardar\", guarda las imágenes en un subdirectorio",
-    "Directory name pattern": "Patrón nombre directorio",
-    "Max prompt words for [prompt_words] pattern": "Máximo de palabras en prompt [prompt_words] para patrón",
-    "Upscaling": "Upscaling",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "Tamaño mosaico para ESRGAN upscalers. 0 = sin mosaico.",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "Solapar mosaico, en pixeles para ESRGAN upscalers. Valores bajos = unión visible.",
-    "Tile size for all SwinIR.": "Tamaño mosaico para SwinIR.",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "Solapar mosaico, en pixeles para SwinIR. Valores bajos = unión visible.",
-    "LDSR processing steps. Lower = faster": "LDSR processing steps. Más bajo = rápido",
-    "Upscaler for img2img": "Upscaler para img2img",
-    "Upscale latent space image when doing hires. fix": "Upscale latent space al aplicar hires. fix",
-    "Face restoration": "Restauración de rostro",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "Parámetro influencia CodeFormer; 0 = máximo efecto; 1 = mínimo efecto",
-    "Move face restoration model from VRAM into RAM after processing": "Mover modelo de restauración de rostro del VRAM al RAM después de procesar",
-    "System": "Sistema",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "Sondeos de uso de VRAM por segundo durante la generación. Establecer en 0 para deshabilitar.",
-    "Always print all generation info to standard output": "Imprime siempre toda la información de generación en la salida estándar",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Agrega una segunda barra de progreso a la consola que muestra el progreso de un trabajo completo.",
-    "Training": "Entrenamiento",
-    "Move VAE and CLIP to RAM when training if possible. Saves VRAM.": "Mover VAE y CLIP al RAM al entrenar cuando sea posible. Ahorra VRAM.",
-    "Filename word regex": "Filename word regex",
-    "Filename join string": "Filename join string",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Número de repeticiones para una sola imagen de entrada por epoch; utilizado solo para mostrar el número epoch",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Guarda un csv que contenga la pérdida en el directorio log cada N pasos, 0 para deshabilitar",
-    "Use cross attention optimizations while training": "Utiliza optimizaciones de atención cruzada durante el entrenamiento",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "Checkpoints al cache en RAM",
-    "SD VAE": "SD VAE",
-    "auto": "auto",
-    "Hypernetwork strength": "Hypernetwork strength",
-    "Inpainting conditioning mask strength": "Fuerza de la máscara en acondicionamiento Inpainting",
-    "Apply color correction to img2img results to match original colors.": "Aplica la corrección de color a los resultados de img2img para que coincidan con los colores originales.",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "Con img2img, hace exactamente la cantidad de pasos que especifica el slider (normalmente haría menos con menos eliminación de ruido).",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Habilita la cuantificación en K samplers para obtener resultados más nítidos y limpios. Esto puede cambiar los seeds existentes. Requiere reiniciar para aplicar.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Énfasis: utiliza (texto) para que el modelo preste más atención al texto y [texto] para que preste menos atención",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Utiliza la implementación de énfasis antiguo. Puede ser útil para reproducir seeds anteriores.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "Hace que los K-diffusion samplers produzcan las mismas imágenes en un lote que cuando se crea una sola imagen",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "Aumenta la coherencia rellenando desde la última coma dentro de n tokens cuando se use más de 75 tokens",
-    "Filter NSFW content": "Filtrar contenido NSFW",
-    "Stop At last layers of CLIP model": "Detener en las últimas capas del modelo CLIP",
-    "Interrogate Options": "Opciones de Interrogar",
-    "Interrogate: keep models in VRAM": "Interrogar: mantener modelos en VRAM",
-    "Interrogate: use artists from artists.csv": "Interrogar: utilizar artistas de artists.csv",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interrogar: incluir rangos de coincidencias de etiquetas de modelos en los resultados (No tiene efecto en los interrogadores basados en subtítulos).",
-    "Interrogate: num_beams for BLIP": "Interrogar: num_beams para BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Interrogar: longitud mínima de la descripción (excluyendo artistas, etc.)",
-    "Interrogate: maximum description length": "Interrogar: longitud máxima de la descripción",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: número máximo de líneas en el archivo de texto (0 = Sin límite)",
-    "Interrogate: deepbooru score threshold": "Interrogar: deepbooru umbral de puntuación",
-    "Interrogate: deepbooru sort alphabetically": "Interrogar: deepbooru ordenar alfabéticamente",
-    "use spaces for tags in deepbooru": "usar espacios para etiquetas en deepbooru",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "corchetes de escape (\\) en deepbooru (por lo que se usan como corchetes literales y no para enfatizar)",
-    "User interface": "Interfaz de usuario",
-    "Show progressbar": "Mostrar barra de progreso",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Muestra el progreso de creación de la imagen cada N sampling steps. Establecer 0 para deshabilitar.",
-    "Show previews of all images generated in a batch as a grid": "Mostrar vistas previas de todas las imágenes generadas en un batch como un grid",
-    "Show grid in results for web": "Mostrar grids en resultados para web",
-    "Do not show any images in results for web": "No mostrar ninguna imagen en los resultados para web",
-    "Add model hash to generation information": "Agregar hash de modelo a la información de generación",
-    "Add model name to generation information": "Agregar nombre de modelo a la información de generación",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "Al leer los parámetros de generación del texto en la interfaz de usuario (desde PNG Info o texto pegado), no cambia el modelo/checkpoint seleccionado.",
-    "Send seed when sending prompt or image to other interface": "Enviar seed cuando se envíe el prompt o imagen a otra interfase",
-    "Font for image grids that have text": "Tipografía para grids de imágenes que tienen texto",
-    "Enable full page image viewer": "Habilitar visor de imágenes de página completa",
-    "Show images zoomed in by default in full page image viewer": "Mostrar imágenes ampliadas de forma predeterminada en el visor de imágenes de página completa",
-    "Show generation progress in window title.": "Muestra el progreso de la generación en el título de la ventana del navegador.",
-    "Quicksettings list": "Lista de ajustes rápidos",
-    "Localization (requires restart)": "Traducción (requiere reiniciar)",
-    "fr_FR": "fr_FR",
-    "tr_TR": "tr_TR",
-    "it_IT": "it_IT",
-    "de_DE": "de_DE",
-    "ru_RU": "ru_RU",
-    "ja_JP": "ja_JP",
-    "es_ES": "es_ES",
-    "ko_KR": "ko_KR",
-    "zh_TW": "zh_TW",
-    "zh_CN": "zh_CN",
-    "ar_AR": "ar_AR",
-    "pt_BR": "pt_BR",
-    "Sampler parameters": "Parámetros del sampler",
-    "Hide samplers in user interface (requires restart)": "Ocultar samplers en interfaz de usuario (requiere reiniciar)",
-    "eta (noise multiplier) for DDIM": "eta (noise multiplier) para DDIM",
-    "eta (noise multiplier) for ancestral samplers": "eta (noise multiplier) para ancestral samplers",
-    "img2img DDIM discretize": "img2img DDIM discretize",
-    "uniform": "uniform",
-    "quad": "quad",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma noise",
-    "Eta noise seed delta": "Eta noise seed delta",
-    "Images Browser": "Navegador de Imágenes",
-    "Preload images at startup": "Precargar imágenes al iniciar",
-    "Number of columns on the page": "Número de columnas en la página",
-    "Number of rows on the page": "Número de filas en la página",
-    "Minimum number of pages per load": "Número mínimo de páginas por carga",
-    "Request browser notifications": "Solicitar notificaciones del navegador",
-    "Download localization template": "Descargar plantilla de traducción",
-    "Reload custom script bodies (No ui updates, No restart)": "Recargar custom script bodies (Sin actualizar UI, Sin reiniciar)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Reiniciar Gradio y Refrescar componentes (Custom Scripts, ui.py, js y css)",
-    "Installed": "Instaladas",
-    "Available": "Disponibles",
-    "Install from URL": "Instalar desde URL",
-    "Apply and restart UI": "Aplicar y reiniciar UI",
-    "Check for updates": "Buscar actualizaciones",
-    "Extension": "Extensión",
-    "URL": "URL",
-    "Update": "Actualizar",
-    "deforum-for-automatic1111-webui": "deforum-for-automatic1111-webui",
-    "https://github.com/deforum-art/deforum-for-automatic1111-webui": "https://github.com/deforum-art/deforum-for-automatic1111-webui",
-    "unknown": "desconocido",
-    "stable-diffusion-webui-images-browser": "stable-diffusion-webui-images-browser",
-    "https://github.com/yfszzx/stable-diffusion-webui-images-browser": "https://github.com/yfszzx/stable-diffusion-webui-images-browser",
-    "Load from:": "Cargar desde:",
-    "Extension index URL": "URL índice de extensiones",
-    "URL for extension's git repository": "URL repositorio git de extensión",
-    "Local directory name": "Nombre directorio local",
-    "Install": "Instalar",
-    "Ver": "Ver",
-    "Entrenar un embedding o Hypernetwork; debes especificar un directorio con un conjunto de imágenes con una proporción de 1:1": "Entrenar un embedding o Hypernetwork; debes especificar un directorio con un conjunto de imágenes con una proporción de 1:1",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "Prompt (presiona Ctrl+Enter o Alt+Enter para generar)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Prompt negativo (presiona Ctrl+Enter o Alt+Enter para generar)",
-    "Add a random artist to the prompt.": "Agregar un artista aleatorio al prompt.",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Leer los parámetros de generación del prompt o de la última generación si el prompt está vacío en la interfaz de usuario.",
-    "Save style": "Guardar estilo",
-    "Apply selected styles to current prompt": "Aplicar estilos seleccionados al prompt",
-    "Stop processing current image and continue processing.": "Dejar de procesar la imagen actual y continuar procesando.",
-    "Stop processing images and return any results accumulated so far.": "Dejar de procesar imágenes y devuelva los resultados acumulados hasta el momento.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Estilo a aplicar; los estilos tienen componentes tanto para prompts positivos como negativos y se aplican a ambos",
-    "Do not do anything special": "No hacer nada especial",
-    "Which algorithm to use to produce the image": "Qué algoritmo usar para producir la imagen",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - muy creativo, cada uno puede obtener una imagen completamente diferente dependiendo del conteo de steps, configurar los steps a más de 30-40 no ayuda",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Models - funcionan mejor para inpainting",
-    "Produce an image that can be tiled.": "Produce una imagen que puede usarse como mosaico.",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Usa un proceso de dos pasos para crear parcialmente una imagen con una resolución más pequeña, mejora la escala y luego mejora los detalles sin cambiar la composición",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Determina qué tan poco respeto debe tener el algoritmo por el contenido de la imagen. En 0, nada cambiará y en 1 obtendrá una imagen no relacionada. Con valores por debajo de 1.0, el procesamiento tomará menos steps de los que especifica el slider de Sampling Steps.",
-    "How many batches of images to create": "Cuantos batches de imágenes para crear",
-    "How many image to create in a single batch": "Cuantas imágenes para crear en un solo batch",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - con qué fuerza debe ajustarse la imagen al prompt: los valores más bajos producen resultados más creativos",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Un valor que determina la salida del generador de números aleatorios: si creas una imagen con los mismos parámetros y el seed que otra imagen, obtendrás el mismo resultado",
-    "Set seed to -1, which will cause a new random number to be used every time": "Establece el seed a -1, lo que hará que se use un nuevo número aleatorio cada vez",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Reutilice el seed de la última generación, muy útil si fue aleatoria",
-    "Seed of a different picture to be mixed into the generation.": "Seed de una imagen diferente para ser mezclada en la generación.",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Qué fuerte de una variación para producir. En 0, no habrá ningún efecto. En 1, obtendrá la imagen completa con variation seed (excepto para ancestral samplers, donde solo obtendrás algo).",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Intenta producir una imagen similar a la que se habría producido con el mismo seed a la resolución especificada",
-    "Separate values for X axis using commas.": "Separar valores para X usando comas.",
-    "Separate values for Y axis using commas.": "Separar valores para Y usando comas.",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Escribe la imagen en un directorio (predeterminado: log/images) y los parámetros de generación en el archivo csv.",
-    "Open images output directory": "Abrir directorio de imágenes",
-    "How much to blur the mask before processing, in pixels.": "Cuánto difuminado a la máscara antes de procesarla, en píxeles.",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Qué poner dentro del área enmascarada antes de procesarla con Stable Diffusion.",
-    "fill it with colors of the image": "rellenarlo con los colores de la imagen",
-    "keep whatever was there originally": "mantener lo que estaba allí originalmente",
-    "fill it with latent space noise": "rellenarlo con latent space noise",
-    "fill it with latent space zeroes": "rellenarlo con latent space zeroes",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "Escalar la región con máscara a la resolución objetivo, vuelve a pintar, reduce la escala hacia atrás y pégala en la imagen original",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Cambia el tamaño de la imagen a la resolución objetivo. A menos que la altura y el ancho coincidan, obtendrás una relación de aspecto incorrecta.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Cambia el tamaño de la imagen para que la totalidad de la resolución destino se llene con la imagen. Recorta las partes que sobresalen.",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Cambia el tamaño de la imagen para que la totalidad de la imagen esté dentro de la resolución objetivo. Rellena el espacio vacío con los colores de la imagen.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Cuántas veces repetir el procesamiento de una imagen y usarla como entrada para la próxima iteración",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "En modo loopback, en cada bucle, la fuerza de eliminación de ruido se multiplica por este valor. <1 significa variedad decreciente, por lo que su secuencia convergerá en una imagen fija. >1 significa aumentar la variedad, por lo que su secuencia se volverá cada vez más caótica.",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "Para SD upscale, cuánta superposición en píxeles debe haber entre mosaicos. Los mosaicos se superponen de modo que cuando se fusionan nuevamente en una imagen, no hay una unión claramente visible.",
-    "A directory on the same machine where the server is running.": "Un directorio en la misma máquina donde se ejecuta el servidor.",
-    "Leave blank to save images to the default path.": "Déjalo en blanco para guardar las imágenes en la ruta predeterminada.",
-    "Result = A * (1 - M) + B * M": "Resultado = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Resultado = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "Primer y último dígito debe ser 1. ej:'1, 2, 1'",
-    "Path to directory with input images": "Ruta al directorio con imágenes de entrada",
-    "Path to directory where to write outputs": "Ruta al directorio donde escribir salidas",
-    "Input images directory": "Directorio de imágenes de entrada",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Usa las siguientes etiquetas para definir cómo se eligen los nombres de archivo para las imágenes: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; dejar vacío para utilizar predeterminados.",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Si esta opción está habilitada, el watermark no se agregará a las imágenes creadas. Advertencia: si no agregas un watermark, es posible que te estés comportando de manera poco ética.",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Usa las siguiente etiquetas para definir cómo los subdirectorios para imágenes y grids son seleccionados: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; dejar vacío para utilizar predeterminados.",
-    "Restore low quality faces using GFPGAN neural network": "Restaurar rostros de baja calidad utilizando GFPGAN neural network",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Esta expresión regular se usará para extraer palabras del nombre de archivo y se unirán usando la opción a continuación en el texto de la etiqueta que se usa para el entrenamiento. Dejar vacío para mantener el texto del nombre de archivo tal como está.",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Esta cadena se usará para unir palabras divididas en una sola línea si la opción anterior está habilitada.",
-    "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "Solo se aplica a modelos inpainting. Determina con qué fuerza enmascarar la imagen original para inpainting en img2img. 1.0 significa totalmente enmascarado, que es el comportamiento predeterminado. 0.0 significa un condicionamiento totalmente desenmascarado. Los valores más bajos ayudarán a preservar la composición general de la imagen, pero tendrán problemas con los grandes cambios.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Lista de nombres de configuración, separados por comas, para configuraciones que deben ir a la barra de acceso rápido en la parte superior, en lugar de la pestaña de configuración habitual. Ver modules/shared.py para configurar los nombres. Requiere reiniciar para aplicar.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Si este valor no es cero, se agregará al seed y se usará para inicializar RNG para ruidos cuando se usan samplers con Eta. Puedes usar esto para producir aún más variaciones de imágenes, o puedes usar esto para hacer coincidir imágenes de otro software si sabes lo que estás haciendo.",
-    "Leave empty for auto": "Dejar vacío para automático"
-}

localizations/fr_FR.json

@@ -1,415 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "Chargement...",
-    "view": "vue",
-    "api": "api",
-    "•": "•",
-    "built with gradio": "Construit avec Gradio",
-    "Stable Diffusion checkpoint": "checkpoint Stable Diffusion",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "Extras",
-    "PNG Info": "Infos PNG",
-    "History": "Historique",
-    "Checkpoint Merger": "Fusion de checkpoints",
-    "Train": "Entrainer",
-    "Settings": "Paramètres",
-    "Prompt": "Requête",
-    "Negative prompt": "Requête négative",
-    "Run": "Lancer",
-    "Skip": "Passer",
-    "Interrupt": "Interrrompre",
-    "Generate": "Générer",
-    "Style 1": "Style 1",
-    "Style 2": "Style 2",
-    "Label": "Etiquette",
-    "File": "Fichier",
-    "Drop File Here": "Déposer votre fichier ici",
-    "-": "-",
-    "or": "ou",
-    "Click to Upload": "Cliquer pour uploader",
-    "Image": "Image",
-    "Check progress": "Voir l'avancement",
-    "Check progress (first)": "Voir l'avancement (1er)",
-    "Sampling Steps": "Étapes d'échantillonnage",
-    "Sampling method": "Méthode d'échantillonnage",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Largeur",
-    "Height": "Hauteur",
-    "Restore faces": "Restaurer les visages",
-    "Tiling": "Mode Tuile",
-    "Highres. fix": "Correction haute résolution",
-    "Firstpass width": "Largeur première passe",
-    "Firstpass height": "Hauteur seconde passe",
-    "Denoising strength": "Puissance de réduction du bruit",
-    "Batch count": "Nombre de lots",
-    "Batch size": "Taille de lots",
-    "CFG Scale": "Echelle CFG",
-    "Seed": "Valeur aléatoire",
-    "Extra": "Extra",
-    "Variation seed": "Variation de la valeur aléatoire",
-    "Variation strength": "Puissance de variation",
-    "Resize seed from width": "Largeur de redimensionnement de la valeur aléatoire",
-    "Resize seed from height": "Hauteur de redimensionnement de la valeur aléatoire",
-    "Script": "Script",
-    "None": "Aucun",
-    "Prompt matrix": "Matrice de requète",
-    "Prompts from file or textbox": "Requètes depuis un fichier ou une boite de dialogue",
-    "X/Y plot": "graphe X/Y",
-    "Put variable parts at start of prompt": "Mettre les mots clés variable au début de la requête",
-    "Show Textbox": "Afficher le champs texte",
-    "File with inputs": "Fichier d'entrée",
-    "Prompts": "Requêtes",
-    "X type": "Paramètre axe X",
-    "Nothing": "Rien",
-    "Var. seed": "Valeur aléatoire variable",
-    "Var. strength": "Puissance variable",
-    "Steps": "Étapes",
-    "Prompt S/R": "Cherche et remplace dans la requête",
-    "Prompt order": "Ordre de la requête",
-    "Sampler": "Echantilloneur",
-    "Checkpoint name": "Nom du checkpoint",
-    "Hypernetwork": "Hypernetwork",
-    "Hypernet str.": "Force de l'Hypernetwork",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma min.",
-    "Sigma max": "Sigma max.",
-    "Sigma noise": "Bruit Sigma",
-    "Eta": "Temps estimé",
-    "Clip skip": "Passer Clip",
-    "Denoising": "Réduction du bruit",
-    "X values": "Valeurs X",
-    "Y type": "Paramètre axe Y",
-    "Y values": "Valeurs Y",
-    "Draw legend": "Afficher la légende",
-    "Include Separate Images": "Inclure les images séparées",
-    "Keep -1 for seeds": "Conserver -1 pour la valeur aléatoire",
-    "Drop Image Here": "Déposer l'image ici",
-    "Save": "Enregistrer",
-    "Send to img2img": "Envoyer vers img2img",
-    "Send to inpaint": "Envoyer vers inpaint",
-    "Send to extras": "Envoyer vers extras",
-    "Make Zip when Save?": "Créer un zip lors de l'enregistrement?",
-    "Textbox": "Champ texte",
-    "Interrogate\nCLIP": "Interroger\nCLIP",
-    "Interrogate\nDeepBooru": "Interroger\nDeepBooru",
-    "Inpaint": "Inpaint",
-    "Batch img2img": "Lot img2img",
-    "Image for img2img": "Image pour img2img",
-    "Image for inpainting with mask": "Image pour inpainting avec masque",
-    "Mask": "Masque",
-    "Mask blur": "Flou masque",
-    "Mask mode": "Mode masque",
-    "Draw mask": "Dessiner masque",
-    "Upload mask": "Uploader masque",
-    "Masking mode": "Mode de masquage",
-    "Inpaint masked": "Inpaint masqué",
-    "Inpaint not masked": "Inpaint non masqué",
-    "Masked content": "Contenu masqué",
-    "fill": "remplir",
-    "original": "original",
-    "latent noise": "bruit latent",
-    "latent nothing": "latent vide",
-    "Inpaint at full resolution": "Inpaint en pleine résolution",
-    "Inpaint at full resolution padding, pixels": "Padding de l'inpaint en pleine résolution, en pixels",
-    "Process images in a directory on the same machine where the server is running.": "Traite les images dans un dossier sur la même machine où le serveur tourne",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Utiliser un dossier de sortie vide pour enregistrer les images normalement plutôt que d'écrire dans le dossier de sortie",
-    "Input directory": "Dossier d'entrée",
-    "Output directory": "Dossier de sortie",
-    "Resize mode": "Mode redimensionnement",
-    "Just resize": "Redimensionner uniquement",
-    "Crop and resize": "Recadrer et redimensionner",
-    "Resize and fill": "Redimensionner et remplir",
-    "img2img alternative test": "Test alternatif img2img",
-    "Loopback": "Bouclage",
-    "Outpainting mk2": "Outpainting v2",
-    "Poor man's outpainting": "Outpainting du pauvre",
-    "SD upscale": "Agrandissement SD",
-    "should be 2 or lower.": "doit être inférieur ou égal à 2",
-    "Override `Sampling method` to Euler?(this method is built for it)": "Forcer `Méthode d'échantillonnage` à Euler ? (cette méthode est dédiée à cela)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "Forcer la `requête` au contenu de la `requête d'origine` ? (de même pour la `requête négative`)",
-    "Original prompt": "Requête d'origine",
-    "Original negative prompt": "Requête négative d'origine",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "Forcer le valeur d'`Étapes d'échantillonnage` à la même valeur qu'`Étapes de décodage` ?",
-    "Decode steps": "Étapes de décodage",
-    "Override `Denoising strength` to 1?": "Forcer `Puissance de réduction du bruit` à 1 ?",
-    "Decode CFG scale": "Echelle CFG de décodage",
-    "Randomness": "Aléatoire",
-    "Sigma adjustment for finding noise for image": "Ajustement Sigma lors de la recherche du bruit dans l'image",
-    "Loops": "Boucles",
-    "Denoising strength change factor": "Facteur de changement de la puissance de réduction du bruit",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Paramètres recommandés : Étapes d'échantillonnage : 80-100, Echantillonneur : Euler a, Puissance de réduction du bruit : 0.8",
-    "Pixels to expand": "Pixels à étendre",
-    "Outpainting direction": "Direction de l'outpainting",
-    "left": "gauche",
-    "right": "droite",
-    "up": "haut",
-    "down": "bas",
-    "Fall-off exponent (lower=higher detail)": "Exposant de diminution (plus petit = plus de détails)",
-    "Color variation": "Variation de couleur",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Agrandira l'image à deux fois sa taille; utilisez les glissières largeur et hauteur afin de choisir la taille de tuile",
-    "Tile overlap": "Chevauchement de tuile",
-    "Upscaler": "Agrandisseur",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "BSRGAN 4x": "BSRGAN 4x",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR 4x": "SwinIR 4x",
-    "Single Image": "Image unique",
-    "Batch Process": "Traitement par lot",
-    "Batch from Directory": "Lot depuis un dossier",
-    "Source": "Source",
-    "Show result images": "Montrez les images résultantes",
-    "Scale by": "Mise à l'échelle de",
-    "Scale to": "Mise à l'échelle à",
-    "Resize": "Redimensionner",
-    "Crop to fit": "Recadrer à la taille",
-    "Upscaler 2": "Agrandisseur 2",
-    "Upscaler 2 visibility": "Visibilité de l'agrandisseur 2",
-    "GFPGAN visibility": "Visibilité GFPGAN",
-    "CodeFormer visibility": "Visibilité CodeFormer",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "Poids CodeFormer (0 = effet maximum, 1 = effet minimum)",
-    "Open output directory": "Ouvrir le dossier de sortie",
-    "Send to txt2img": "Envoyer vers txt2img",
-    "txt2img history": "historique txt2img",
-    "img2img history": "historique img2img",
-    "extras history": "historique extras",
-    "Renew Page": "Rafraichr la page",
-    "First Page": "Première page",
-    "Prev Page": "Page précendente",
-    "Page Index": "Index des pages",
-    "Next Page": "Page suivante",
-    "End Page": "Page de fin",
-    "number of images to delete consecutively next": "nombre d'image à supprimer consécutivement ensuite",
-    "Delete": "Supprimer",
-    "Generate Info": "Générer les informations",
-    "File Name": "Nom de fichier",
-    "set_index": "set_index",
-    "A merger of the two checkpoints will be generated in your": "Une fusion des deux checkpoints sera générée dans votre",
-    "checkpoint": "checkpoint",
-    "directory.": "dossier",
-    "Primary model (A)": "Modèle primaire (A)",
-    "Secondary model (B)": "Modèle secondaire (B)",
-    "Tertiary model (C)": "Modèle tertiaire (C)",
-    "Custom Name (Optional)": "Nom personnalisé (Optionel)",
-    "Multiplier (M) - set to 0 to get model A": "Multiplieur (M) - utiliser 0 pour le modèle A",
-    "Interpolation Method": "Méthode d'interpolation",
-    "Weighted sum": "Somme pondérée",
-    "Add difference": "Ajouter différence",
-    "Save as float16": "Enregistrer en tant que float16",
-    "See": "Voir",
-    "wiki": "wiki",
-    "for detailed explanation.": "pour une explication détaillée.",
-    "Create embedding": "Créer un embedding",
-    "Create hypernetwork": "Créer un hypernetwork",
-    "Preprocess images": "Pré-traite les images",
-    "Name": "Nom",
-    "Initialization text": "Texte d'initialisation",
-    "Number of vectors per token": "Nombre de vecteurs par jeton",
-    "Modules": "Modules",
-    "Source directory": "Dossier source",
-    "Destination directory": "Dossier destination",
-    "Create flipped copies": "Créer des copies en mirroir",
-    "Split oversized images into two": "Couper les images trop grandes en deux",
-    "Use BLIP for caption": "Utiliser BLIP pour les descriptions",
-    "Use deepbooru for caption": "Utiliser deepbooru pour les descriptions",
-    "Preprocess": "Pré-traite",
-    "Train an embedding; must specify a directory with a set of 1:1 ratio images": "Entrainer un embedding ; spécifiez un dossier contenant un ensemble d'images avec un ratio de 1:1",
-    "Embedding": "Embedding",
-    "Learning rate": "Vitesse d'apprentissage",
-    "Dataset directory": "Dossier des images d'entrée",
-    "Log directory": "Dossier de journalisation",
-    "Prompt template file": "Fichier modèle de requêtes",
-    "Max steps": "Étapes max.",
-    "Save an image to log directory every N steps, 0 to disable": "Enregistrer une image dans le dossier de journalisation toutes les N étapes, 0 pour désactiver",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Enregistrer une copie de l'embedding dans le dossier de journalisation toutes les N étapes, 0 pour désactiver",
-    "Save images with embedding in PNG chunks": "Sauvegarder les images incluant l'embedding dans leur blocs PNG",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Lire les paramètres (requête, etc.) depuis l'onglet txt2img lors de la génération des previews",
-    "Train Hypernetwork": "Entrainer un Hypernetwork",
-    "Train Embedding": "Entrainer un Embedding",
-    "Apply settings": "Appliquer les paramètres",
-    "Saving images/grids": "Enregistrer les images/grilles",
-    "Always save all generated images": "Toujours enregistrer toutes les images",
-    "File format for images": "Format de fichier pour les images",
-    "Images filename pattern": "Motif pour le nom de fichier des images",
-    "Always save all generated image grids": "Toujours enregistrer toutes les grilles d'images générées",
-    "File format for grids": "Format de fichier pour les grilles",
-    "Add extended info (seed, prompt) to filename when saving grid": "Ajouter les informations étendues (valeur aléatoire, requête) aux noms de fichiers lors de l'enregistrement d'une grille",
-    "Do not save grids consisting of one picture": "Ne pas enregistrer les grilles contenant une seule image",
-    "Prevent empty spots in grid (when set to autodetect)": "Eviter les vides dans la grille (quand autodétection est choisie)",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Nombre de colonnes de la grille; utilisez -1 pour autodétection et 0 pour qu'il soit égal à la taille du lot",
-    "Save text information about generation parameters as chunks to png files": "Enregistrer l'information du text des paramètres de génération en tant que blocs dans les fichiers PNG",
-    "Create a text file next to every image with generation parameters.": "Créer un fichier texte contenant les paramètres de génération à côté de chaque image",
-    "Save a copy of image before doing face restoration.": "Enregistrer une copie de l'image avant de lancer la restauration de visage",
-    "Quality for saved jpeg images": "Qualité pour les images jpeg enregistrées",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "Si l'image PNG est plus grande que 4MB or l'une des ses dimensions supérieure à 4000, réduire sa taille et enregistrer une copie en JPG",
-    "Use original name for output filename during batch process in extras tab": "Utiliser un nom de fichier original pour les fichiers de sortie durant le traitement par lot dans l'onglet Extras",
-    "When using 'Save' button, only save a single selected image": "A l'utilisation du bouton `Enregistrer`, n'enregistrer que l'image séléctionnée",
-    "Do not add watermark to images": "Ne pas ajouter de filigrane aux images",
-    "Paths for saving": "Chemins pour l'enregistrement",
-    "Output directory for images; if empty, defaults to three directories below": "Dossier de sortie pour les images; si non spécifié, le chemin par défaut sera trois niveau en dessous",
-    "Output directory for txt2img images": "Dossier de sortie pour les images txt2img",
-    "Output directory for img2img images": "Dossier de sortie pour les images img2img",
-    "Output directory for images from extras tab": "Dossier de sortie pour les images de l'onglet Extras",
-    "Output directory for grids; if empty, defaults to two directories below": "Dossier de sortie pour les grilles; si non spécifié, le chemin par défaut sera deux niveau en dessous",
-    "Output directory for txt2img grids": "Dossier de sortie pour les grilles txt2img",
-    "Output directory for img2img grids": "Dossier de sortie pour les grilles img2img",
-    "Directory for saving images using the Save button": "Dossier de sauvegarde des images pour le bouton `Enregistrer`",
-    "Saving to a directory": "Enregistrer dans un dossier",
-    "Save images to a subdirectory": "Enregistrer les images dans un sous dossier",
-    "Save grids to a subdirectory": "Enregistrer les grilles dans un sous dossier",
-    "When using \"Save\" button, save images to a subdirectory": "Lors de l'utilisation du bouton \"Enregistrer\", sauvegarder les images dans un sous dossier",
-    "Directory name pattern": "Motif pour le nom des dossiers",
-    "Max prompt words for [prompt_words] pattern": "Maximum de mot pour le motif [prompt_words]",
-    "Upscaling": "Agrandissement",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "Taille des tuile for les agrandisseurs ESRGAN. 0 = mode tuile désactivé.",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "Chevauchement des tuiles, en pixel pour l'agrandisseur ESRGAN. Valeur faible = couture visible",
-    "Tile size for all SwinIR.": "Taille de la tuile pour tous les agrandisseur SwinIR.",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "Chevauchement de tuile, en pixels pour SwinIR. Valeur faible = couture visible",
-    "LDSR processing steps. Lower = faster": "Echantillon du traitement LDSR. Valeur faible =  plus rapide",
-    "Upscaler for img2img": "Agrandisseur pour img2img",
-    "Upscale latent space image when doing hires. fix": "Agrandir l'image de l'espace latent lors de la correction haute résolution",
-    "Face restoration": "Restauration de visage",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "Paramètre de poids pour CodeFormer; 0 = effet maximum ； 1 = effet minimum",
-    "Move face restoration model from VRAM into RAM after processing": "Déplacer le modèle de restauration de visage de la VRAM vers la RAM après traitement",
-    "System": "Système",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "Fréquence d'interrogation par seconde pendant la génération. Mettez la valeur à 0 pour désactiver.",
-    "Always print all generation info to standard output": "Toujours afficher toutes les informations de génération dans la sortie standard",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Ajouter un seconde barre de progression dans la console montrant l'avancement pour un tâche complète.",
-    "Training": "Entrainement",
-    "Unload VAE and CLIP from VRAM when training": "Décharger VAE et CLIP de la VRAM pendant l'entrainement",
-    "Filename word regex": "Regex de mot",
-    "Filename join string": "Chaine de caractère pour lier les noms de fichier",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Nombre de répétition pour une image unique par époque; utilisé seulement pour afficher le nombre d'époques",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Enregistrer un csv contenant la perte dans le dossier de journalisation toutes les N étapes, 0 pour désactiver",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "Checkpoint à mettre en cache dans la RAM",
-    "Hypernetwork strength": "Force de l'Hypernetwork",
-    "Apply color correction to img2img results to match original colors.": "Appliquer une correction de couleur aux résultats img2img afin de conserver les couleurs d'origine",
-    "Save a copy of image before applying color correction to img2img results": "Enregistrer une copie de l'image avant d'appliquer les résultats de la correction de couleur img2img",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "Avec img2img, executer exactement le nombre d'étapes spécifiées par la glissière (normalement moins d'étapes sont executées quand la réduction du bruit est plus faible).",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Activer la quantisation des échantillionneurs K pour des résultats plus nets et plus propres. Cela peut modifier les valeurs aléatoires existantes. Requiert un redémarrage pour être actif.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Emphase : utilisez (texte) afin de forcer le modèle à porter plus d'attention au texte et [texte] afin qu'il y porte moins attention",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Utilisez l'ancienne méthode d'emphase. Peut être utile afin de reproduire d'anciennes valeurs aléatoires.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "Demander aux échantillionneurs K-diffusion de produire les mêmes dans un lot que lors de la génération d'une image unique",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "Améliorer la cohérence en remplissant (padding) à partir de la dernière virgule dans les X jetons quand on en utilise plus de 75",
-    "Filter NSFW content": "Filtrer le contenu +18 (NSFW)",
-    "Stop At last layers of CLIP model": "S'arrêter aux derniers niveaux du modèle CLIP",
-    "Interrogate Options": "Options d'intérrogation",
-    "Interrogate: keep models in VRAM": "Interroger : conserver les modèles en VRAM",
-    "Interrogate: use artists from artists.csv": "Interroger : utiliser les artistes dans artists.csv",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interroger : inclure la correspondance du classement des labels de modèle dans les résultats (N'a pas d'effet sur les interrogateurs basés sur des descriptions) ",
-    "Interrogate: num_beams for BLIP": "Interroger : num_beams pour BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Interroger : longueur minimale de la description (excluant les artistes, etc.)",
-    "Interrogate: maximum description length": "Interroger : longueur maximale de la description",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP : nombre maximum de lignes dans le fichier texte (0 = pas de limite)",
-    "Interrogate: deepbooru score threshold": "Interroger : seuil du score deepbooru",
-    "Interrogate: deepbooru sort alphabetically": "Interroger : classement alphabétique deepbooru",
-    "use spaces for tags in deepbooru": "Utiliser des espaces pour les étiquettes dans deepbooru",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "échapper (\\) les crochets dans deepbooru (afin qu'ils puissent être utilisés littéralement et non pour mettre en emphase)",
-    "User interface": "Interface utilisateur",
-    "Show progressbar": "Afficher la barre de progression",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Afficher l'état d'avancement de la création d'image toutes les X étapes d'échantillionnage. Utiliser 0 pour désactiver.",
-    "Show grid in results for web": "Afficher la grille dans les résultats web",
-    "Do not show any images in results for web": "N'afficher aucune image dans les résultats web'",
-    "Add model hash to generation information": "Ajouter le hash du modèle dans l'information de génération",
-    "Add model name to generation information": "Ajouter le nom du modèle dans l'information de génération",
-    "Font for image grids that have text": "Police pour les grilles d'images contenant du texte",
-    "Enable full page image viewer": "Activer l'affichage des images en plein écran",
-    "Show images zoomed in by default in full page image viewer": "Afficher les images zoomées par défaut lors de l'affichage en plein écran",
-    "Show generation progress in window title.": "Afficher l'avancement de la génération dans le titre de la fenêtre.",
-    "Quicksettings list": "Liste de réglages rapides",
-    "Localization (requires restart)": "Localisation (requiert un redémarrage)",
-    "Sampler parameters": "Paramètres de l'échantillionneur",
-    "Hide samplers in user interface (requires restart)": "Cacher les échantillonneurs dans l'interface utilisateur (requiert un redémarrage)",
-    "eta (noise multiplier) for DDIM": "eta (multiplicateur de bruit) pour DDIM",
-    "eta (noise multiplier) for ancestral samplers": "eta (multiplicateur de bruit) poru les échantillionneurs de type 'ancestral'",
-    "img2img DDIM discretize": "Discrétisation DDIM pour img2img",
-    "uniform": "uniforme",
-    "quad": "quad",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma noise",
-    "Eta noise seed delta": "Eta noise seed delta",
-    "Request browser notifications": "Demander les notifications au navigateur",
-    "Download localization template": "Télécharger le modèle de localisation",
-    "Reload custom script bodies (No ui updates, No restart)": "Recharger le contenu des scripts personnalisés (Pas de mise à jour de l'interface, Pas de redémarrage)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Redémarrer Gradio et rafraichir les composants (Scripts personnalisés, ui.py, js et css uniquement)",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "Requête (Ctrl + Entrée ou Alt + Entrée pour générer)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Requête négative (Ctrl + Entrée ou Alt + Entrée pour générer)",
-    "Add a random artist to the prompt.": "Ajouter un artiste aléatoire à la requête",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Lire les paramètres de génération depuis la requête, ou depuis la dernière génération si la requête est vide dans l'interface utilisateur.",
-    "Save style": "Sauvegarder le style",
-    "Apply selected styles to current prompt": "Appliquer les styles séléctionnés à la requête actuelle",
-    "Stop processing current image and continue processing.": "Arrêter le traitement de l'image actuelle et continuer le traitement.",
-    "Stop processing images and return any results accumulated so far.": "Arrêter le traitement des images et retourne les résultats accumulés depuis le début.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Style à appliquer ; les styles sont composés de requêtes positives et négatives et s'appliquent au deux",
-    "Do not do anything special": "Ne rien faire de particulier",
-    "Which algorithm to use to produce the image": "Quel algorithme utiliser pour produire l'image",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - très créatif, peut générer des images complètement différentes en fonction du nombre d'étapes, utiliser plus de 30 à 40 étapes n'améliore pas le résultat",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Modèles implicite de réduction du bruit à diffusion - utile pour l'inpainting",
-    "Produce an image that can be tiled.": "Produit une image qui peut être bouclée (tuile).",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Utilise un processus en deux étapes afin de créer partiellement une image dans une résolution plus faible, l'agrandir et améliorer ses détails sans modifier la composition",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Détermine à quel point l'algorithme doit respecter le contenu de l'image. A 0 rien ne changera, à 1 l'image sera entièrement différente. Avec des valeurs inférieures à 1.0 le traitement utilisera moins d'étapes que ce que la glissière Étapes d'échantillionnage spécifie. ",
-    "How many batches of images to create": "Combien de lots d'images créer",
-    "How many image to create in a single batch": "Combien d'images créer par lot",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - spécifie à quel point l'image doit se conformer à la requête - des valeurs plus faibles produisent des résultats plus créatifs",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Une valeur qui détermine la sortie du générateur de nombres aléatoires - si vous créez une image avec les mêmes paramètres et valeur aléatoire qu'une autre, le résultat sera identique",
-    "Set seed to -1, which will cause a new random number to be used every time": "Passer la valeur aléatoire à -1, cela causera qu'un nombre aléatoire différent sera utilisé à chaque fois",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Réutiliser la valeur aléatoire de la dernière génération, généralement utile uniquement si elle était randomisée",
-    "Seed of a different picture to be mixed into the generation.": "Valeur aléatoire d'une image différente à mélanger dans la génération",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Force de la variation à produire. A 0 il n'y aura pas d'effet. A 1 l'image sera composée uniquement de la valeur aléatoire variable spécifiée (à l'exception des échantillionneurs `ancestral`)",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Essayer de produire une image similaire à ce qu'elle aurait été avec la même valeur aléatoire, mais dans la résolution spécifiée",
-    "Separate values for X axis using commas.": "Séparer les valeurs pour l'axe X par des virgules",
-    "Separate values for Y axis using commas.": "Séparer les valeurs pour l'axe Y par des virgules",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Ecrire l'image dans un dossier (par défaut - log/images) et les paramètres de génération dans un fichier csv.",
-    "Open images output directory": "Ouvrir le dossier de sortie des images",
-    "How much to blur the mask before processing, in pixels.": "Quantité de flou à appliquer au masque avant traitement, en pixels",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Avec quoi remplir la zone masquée avant traitement par Stable Diffusion.",
-    "fill it with colors of the image": "remplir avec les couleurs de l'image",
-    "keep whatever was there originally": "conserver ce qui était présent à l'origine",
-    "fill it with latent space noise": "remplir avec le bruit de l'espace latent",
-    "fill it with latent space zeroes": "remplir avec des zéros dans l'espace latent",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "Agrandir la région masquées à la résolution cible, exécuter l'inpainting, réduire à nouveau puis coller dans l'image originale",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Redimensionner l'image dans la résolution cible. A moins que la hauteur et la largeur coincident le ratio de l'image sera incorrect.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Redimensionner l'image afin que l'entièreté de la résolution cible soit remplie par l'image. Recadrer les parties qui dépassent.",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Redimensionner l'image afin que l'entièreté de l'image soit contenue dans la résolution cible. Remplir l'espace vide avec les couleurs de l'image.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Combien de fois répéter le traitement d'une image et l'utiliser comme entrée pour la prochaine itération",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "En mode bouclage (Loopback), à chaque tour de la boucle la force du réducteur de bruit est multipliée par cette valeur. <1 signifie réduire la variation donc votre séquence convergera vers une image fixe. >1 signifie augmenter la variation donc votre séquence deviendra de plus en plus chaotique. ",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "Pour l'agrandissement SD, de combien les tuiles doivent se chevaucher, en pixels. Les tuiles se chevauchent de manière à ce qu'il n'y ait pas de couture visible une fois fusionnées en une image. ",
-    "A directory on the same machine where the server is running.": "Un dossier sur la même machine où le serveur tourne.",
-    "Leave blank to save images to the default path.": "Laisser vide pour sauvegarder les images dans le chemin par défaut.",
-    "Result = A * (1 - M) + B * M": "Résultat = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Résultat = A + (B - C) * M",
-    "Path to directory with input images": "Chemin vers le dossier contenant les images d'entrée",
-    "Path to directory where to write outputs": "Chemin vers le dossier où écrire les sorties",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "Utiliser les étiquettes suivantes pour définir le nom des images : [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp] ; laisser vide pour le nom par défaut.",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Si cette option est activée le filigrane ne sera pas ajouté au images crées. Attention : si vous n'ajoutez pas de filigrane vous pourriez vous comporter de manière non éthique.",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "Utiliser les étiquettes suivantes pour définir le nom des sous dossiers pour les images et les grilles  : [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp] ; laisser vide pour le nom par défaut.",
-    "Restore low quality faces using GFPGAN neural network": "Restaurer les visages de basse qualité en utilisant le réseau neuronal GFPGAN",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Cette expression régulière sera utilisée pour extraire les mots depuis le nom de fichier ; ils seront joints en utilisant l'option ci dessous en une étiquette utilisée pour l'entrainement. Laisser vide pour conserver le texte du nom de fichier tel quel.",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Cette chaine de caractères sera utilisée pour joindre les mots séparés en une ligne unique si l'option ci dessus est activée.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Liste des noms de paramètres, séparés par des virgules, pour les paramètres de la barre d'accès rapide en haut de page, plutôt que dans la page habituelle des paramètres. Voir modules/shared.py pour définir les noms. Requiert un redémarrage pour s'appliquer.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Si cette valeur est différente de zéro elle sera ajoutée à la valeur aléatoire et utilisée pour initialiser le générateur de nombres aléatoires du bruit lors de l'utilisation des échantillonneurs supportants Eta. Vous pouvez l'utiliser pour produire encore plus de variation dans les images, ou vous pouvez utiliser ceci pour faire correspondre les images avec d'autres logiciels si vous savez ce que vous faites.",
-    "Enable Autocomplete": "Activer l'autocomplétion",
-    "/0.0": "/0.0"
-}

localizations/ja_JP.json

@@ -1,482 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "読み込み中...",
-    "view": "view",
-    "api": "api",
-    "•": "•",
-    "gradioで作ろう": "gradioで作ろう",
-    "Stable Diffusion checkpoint": "Stable Diffusion checkpoint",
-    "Stop At last layers of CLIP model": "最後から何層目でCLIPを止めるか",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "その他",
-    "PNG Info": "PNG内の情報を表示",
-    "Image Browser": "画像閲覧",
-    "Checkpoint Merger": "Checkpointの統合",
-    "Train": "学習",
-    "Create aesthetic embedding": "aesthetic embeddingを作る",
-    "Settings": "設定",
-    "Prompt": "プロンプト",
-    "Negative prompt": "ネガティブ プロンプト",
-    "Run": "実行",
-    "Skip": "スキップ",
-    "Interrupt": "中断",
-    "Generate": "生成！",
-    "Style 1": "スタイル 1",
-    "Style 2": "スタイル 2",
-    "Label": "ラベル",
-    "File": "ファイル",
-    "ここにファイルをドロップ": "ここにファイルをドロップ",
-    "-": "-",
-    "または": "または",
-    "クリックしてアップロード": "クリックしてアップロード",
-    "Image": "画像",
-    "Check progress": "Check progress",
-    "Check progress (first)": "Check progress (first)",
-    "Sampling Steps": "サンプリング回数",
-    "Sampling method": "サンプリングアルゴリズム",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "幅",
-    "Height": "高さ",
-    "Restore faces": "顔修復",
-    "Tiling": "テクスチャ生成モード",
-    "Highres. fix": "高解像度 fix(マウスオーバーで詳細)",
-    "Firstpass width": "Firstpass width",
-    "Firstpass height": "Firstpass height",
-    "Denoising strength": "ノイズ除去強度",
-    "Batch count": "バッチ生成回数",
-    "Batch size": "バッチあたり生成枚数",
-    "CFG Scale": "CFG Scale",
-    "Seed": "シード値",
-    "Extra": "その他",
-    "Variation seed": "Variation シード値",
-    "Variation strength": "Variation 強度",
-    "Resize seed from width": "Resize seed from width",
-    "Resize seed from height": "Resize seed from height",
-    "Open for Clip Aesthetic!": "Open for Clip Aesthetic!",
-    "▼": "▼",
-    "Aesthetic weight": "Aesthetic weight",
-    "Aesthetic steps": "Aesthetic steps",
-    "Aesthetic learning rate": "Aesthetic learning rate",
-    "Slerp interpolation": "Slerp interpolation",
-    "Aesthetic imgs embedding": "Aesthetic imgs embedding",
-    "None": "なし",
-    "Aesthetic text for imgs": "Aesthetic text for imgs",
-    "Slerp angle": "Slerp angle",
-    "Is negative text": "Is negative text",
-    "Script": "スクリプト",
-    "Prompt matrix": "Prompt matrix",
-    "Prompts from file or textbox": "Prompts from file or textbox",
-    "Save steps of the sampling process to files": "Save steps of the sampling process to files",
-    "X/Y plot": "X/Y plot",
-    "Put variable parts at start of prompt": "Put variable parts at start of prompt",
-    "Show Textbox": "Show Textbox",
-    "File with inputs": "File with inputs",
-    "Prompts": "プロンプト",
-    "Save images to path": "Save images to path",
-    "X type": "X軸の種類",
-    "Nothing": "なし",
-    "Var. seed": "Var. seed",
-    "Var. strength": "Var. 強度",
-    "Steps": "ステップ数",
-    "Prompt S/R": "Prompt S/R",
-    "Prompt order": "Prompt order",
-    "Sampler": "サンプラー",
-    "Checkpoint name": "Checkpoint名",
-    "Hypernetwork": "Hypernetwork",
-    "Hypernet str.": "Hypernetの強度",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma min",
-    "Sigma max": "Sigma max",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Eta",
-    "Clip skip": "Clip skip",
-    "Denoising": "Denoising",
-    "X values": "Xの値",
-    "Y type": "Y軸の種類",
-    "Y values": "Yの値",
-    "Draw legend": "凡例を描画",
-    "Include Separate Images": "Include Separate Images",
-    "Keep -1 for seeds": "シード値を-1で固定",
-    "ここに画像をドロップ": "ここに画像をドロップ",
-    "Save": "保存",
-    "Send to img2img": "img2imgに送る",
-    "Send to inpaint": "描き直しに送る",
-    "Send to extras": "その他タブに送る",
-    "Make Zip when Save?": "保存するときZipも同時に作る",
-    "Textbox": "Textbox",
-    "Interrogate\nCLIP": "Interrogate\nCLIP",
-    "Interrogate\nDeepBooru": "Interrogate\nDeepBooru",
-    "Inpaint": "描き直し(Inpaint)",
-    "Batch img2img": "Batch img2img",
-    "Image for img2img": "Image for img2img",
-    "Image for inpainting with mask": "Image for inpainting with mask",
-    "Mask": "マスク",
-    "Mask blur": "マスクぼかし",
-    "Mask mode": "マスクモード",
-    "Draw mask": "マスクをかける",
-    "Upload mask": "マスクをアップロードする",
-    "Masking mode": "マスキング方法",
-    "Inpaint masked": "マスクされた場所を描き直す",
-    "Inpaint not masked": "マスクされていない場所を描き直す",
-    "Masked content": "マスクされたコンテンツ",
-    "fill": "埋める",
-    "original": "オリジナル",
-    "latent noise": "潜在空間でのノイズ",
-    "latent nothing": "潜在空間での無",
-    "Inpaint at full resolution": "フル解像度で描き直す",
-    "Inpaint at full resolution padding, pixels": "フル解像度で描き直す際のパディング数。px単位。",
-    "Process images in a directory on the same machine where the server is running.": "サーバーが稼働しているマシンと同じフォルダにある画像を処理します",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "\"出力フォルダ\"を空にすると、通常の画像と同様に保存されます。",
-    "Input directory": "入力フォルダ",
-    "Output directory": "出力フォルダ",
-    "Resize mode": "リサイズモード",
-    "Just resize": "リサイズのみ",
-    "Crop and resize": "切り取ってからリサイズ",
-    "Resize and fill": "リサイズして埋める",
-    "img2img alternative test": "img2img alternative test",
-    "Loopback": "ループバック",
-    "Outpainting mk2": "Outpainting mk2",
-    "Poor man's outpainting": "Poor man's outpainting",
-    "SD upscale": "SD アップスケール",
-    "[C] Video to video": "[C] Video to video",
-    "should be 2 or lower.": "2以下にすること",
-    "Override `Sampling method` to Euler?(this method is built for it)": "サンプリングアルゴリズムをEulerに上書きする(そうすることを前提に設計されています)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "プロンプトをオリジナルプロンプトと同じ値に上書きする(ネガティブプロンプトも同様)",
-    "Original prompt": "オリジナルのプロンプト",
-    "Original negative prompt": "オリジナルのネガティブプロンプト",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "サンプリング数をデコードステップ数と同じ値に上書きする",
-    "Decode steps": "デコードステップ数",
-    "Override `Denoising strength` to 1?": "ノイズ除去強度を1に上書きする",
-    "Decode CFG scale": "Decode CFG scale",
-    "Randomness": "ランダム性",
-    "Sigma adjustment for finding noise for image": "Sigma adjustment for finding noise for image",
-    "Loops": "ループ数",
-    "Denoising strength change factor": "Denoising strength change factor",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "推奨設定: サンプリング回数: 80-100, サンプリングアルゴリズム: Euler a, ノイズ除去強度: 0.8",
-    "Pixels to expand": "Pixels to expand",
-    "Outpainting direction": "Outpainting direction",
-    "left": "左",
-    "right": "右",
-    "up": "上",
-    "down": "下",
-    "Fall-off exponent (lower=higher detail)": "Fall-off exponent (lower=higher detail)",
-    "Color variation": "Color variation",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "画像を2倍の大きさにアップスケールします。幅と高さのスライダーでタイルの大きさを設定します。",
-    "Tile overlap": "Tile overlap",
-    "Upscaler": "アップスケーラー",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "R-ESRGAN General 4xV3": "R-ESRGAN General 4xV3",
-    "R-ESRGAN General WDN 4xV3": "R-ESRGAN General WDN 4xV3",
-    "R-ESRGAN AnimeVideo": "R-ESRGAN AnimeVideo",
-    "R-ESRGAN 4x+": "R-ESRGAN 4x+",
-    "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-    "R-ESRGAN 2x+": "R-ESRGAN 2x+",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR 4x": "SwinIR 4x",
-    "Input file path": "Input file path",
-    "CRF (quality, less is better, x264 param)": "CRF (quality, less is better, x264 param)",
-    "FPS": "FPS",
-    "Seed step size": "Seed step size",
-    "Seed max distance": "Seed max distance",
-    "Start time": "Start time",
-    "End time": "End time",
-    "Single Image": "単一画像",
-    "Batch Process": "バッチ処理",
-    "Batch from Directory": "フォルダからバッチ処理",
-    "Source": "入力",
-    "Show result images": "出力画像を表示",
-    "Scale by": "倍率指定",
-    "Scale to": "解像度指定",
-    "Resize": "倍率",
-    "Crop to fit": "合うように切り抜き",
-    "Upscaler 2 visibility": "Upscaler 2 visibility",
-    "GFPGAN visibility": "GFPGAN visibility",
-    "CodeFormer visibility": "CodeFormer visibility",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormerの重み (注:0で最大、1で最小)",
-    "Open output directory": "出力フォルダを開く",
-    "Send to txt2img": "txt2imgに送る",
-    "extras": "その他タブ",
-    "favorites": "お気に入り",
-    "Load": "読み込み",
-    "Images directory": "フォルダ",
-    "Prev batch": "前の batch",
-    "Next batch": "次の batch",
-    "First Page": "最初のぺージへ",
-    "Prev Page": "前ページへ",
-    "Page Index": "ページ番号",
-    "Next Page": "次ページへ",
-    "End Page": "最後のページへ",
-    "number of images to delete consecutively next": "次の削除で一度に削除する画像数",
-    "Delete": "削除",
-    "Generate Info": "生成情報",
-    "File Name": "ファイル名",
-    "Collect": "保存(お気に入り)",
-    "Refresh page": "ページを更新",
-    "Date to": "Date to",
-    "Number": "Number",
-    "set_index": "set_index",
-    "Checkbox": "Checkbox",
-    "A merger of the two checkpoints will be generated in your": "統合されたチェックポイントはあなたの",
-    "checkpoint": "checkpoint",
-    "directory.": "フォルダに保存されます.",
-    "Primary model (A)": "1つめのmodel (A)",
-    "Secondary model (B)": "2つめのmodel (B)",
-    "Tertiary model (C)": "3つめのmodel (C)",
-    "Custom Name (Optional)": "Custom Name (任意)",
-    "Multiplier (M) - set to 0 to get model A": "Multiplier (M) 0にすると完全にmodel Aとなります (ツールチップ参照)",
-    "Interpolation Method": "混合(Interpolation)方式",
-    "Weighted sum": "加重平均",
-    "Add difference": "差を加える",
-    "Save as float16": "float16で保存",
-    "See": "詳細な説明については",
-    "wiki": "wiki",
-    "for detailed explanation.": "を見てください。",
-    "Create embedding": "Embeddingを作る",
-    "Create hypernetwork": "Hypernetworkを作る",
-    "Preprocess images": "画像の前処理",
-    "Name": "名称",
-    "Initialization text": "Initialization text",
-    "Number of vectors per token": "Number of vectors per token",
-    "Overwrite Old Embedding": "古いEmbeddingを上書き",
-    "Modules": "モジュール",
-    "Enter hypernetwork layer structure": "Hypernetworkのレイヤー構造を入力",
-    "Select activation function of hypernetwork": "Hypernetworkの活性化関数",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "Add layer normalization": "Add layer normalization",
-    "Use dropout": "Use dropout",
-    "Overwrite Old Hypernetwork": "古いHypernetworkを上書きする",
-    "Source directory": "入力フォルダ",
-    "Destination directory": "出力フォルダ",
-    "Existing Caption txt Action": "既存のキャプションの取り扱い",
-    "ignore": "無視する",
-    "copy": "コピーする",
-    "prepend": "先頭に加える",
-    "append": "末尾に加える",
-    "Create flipped copies": "反転画像を生成する",
-    "Split oversized images": "大きすぎる画像を分割する",
-    "Use BLIP for caption": "BLIPで説明をつける",
-    "Use deepbooru for caption": "deepbooruで説明をつける",
-    "Split image threshold": "分割する大きさの閾値",
-    "Split image overlap ratio": "Split image overlap ratio",
-    "Preprocess": "前処理開始",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "EmbeddingまたはHypernetworkを学習します。1:1の比率の画像セットを含むフォルダを指定する必要があります。",
-    "[wiki]": "[wiki]",
-    "Embedding": "Embedding",
-    "Embedding Learning rate": "Embeddingの学習率(Learning rate)",
-    "Hypernetwork Learning rate": "Hypernetworkの学習率(Learning rate)",
-    "Dataset directory": "データセットフォルダ",
-    "Log directory": "ログフォルダ",
-    "Prompt template file": "プロンプトのテンプレートファイル",
-    "Max steps": "最大ステップ数",
-    "Save an image to log directory every N steps, 0 to disable": "指定したステップ数ごとに画像を生成し、ログに保存する。0で無効化。",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "指定したステップ数ごとにEmbeddingのコピーをログに保存する。0で無効化。",
-    "Save images with embedding in PNG chunks": "保存する画像にembeddingを埋め込む",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "プレビューの作成にtxt2imgタブから読み込んだパラメータ(プロンプトなど)を使う",
-    "Train Hypernetwork": "Hypernetworkの学習を開始",
-    "Train Embedding": "Embeddingの学習を開始",
-    "Create an aesthetic embedding out of any number of images": "Create an aesthetic embedding out of any number of images",
-    "Create images embedding": "Create images embedding",
-    "Apply settings": "設定を適用",
-    "Saving images/grids": "画像/グリッドの保存",
-    "Always save all generated images": "生成された画像をすべて保存する",
-    "File format for images": "画像ファイルの保存形式",
-    "Images filename pattern": "ファイル名のパターン",
-    "Always save all generated image grids": "グリッド画像を常に保存する",
-    "File format for grids": "グリッド画像の保存形式",
-    "Add extended info (seed, prompt) to filename when saving grid": "保存するグリッド画像のファイル名に追加情報(シード値、プロンプト)を加える",
-    "Do not save grids consisting of one picture": "1画像からなるグリッド画像は保存しない",
-    "Prevent empty spots in grid (when set to autodetect)": "(自動設定のとき)グリッドに空隙が生じるのを防ぐ",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "グリッドの列数; -1で自動設定、0でバッチ生成回数と同じにする",
-    "Save text information about generation parameters as chunks to png files": "生成に関するパラメーターをPNG画像に含める",
-    "Create a text file next to every image with generation parameters.": "保存する画像とともに生成パラメータをテキストファイルで保存する",
-    "Save a copy of image before doing face restoration.": "顔修復を行う前にコピーを保存しておく。",
-    "Quality for saved jpeg images": "JPG保存時の画質",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "PNG画像が4MBを超えるか、どちらか1辺の長さが4000を超えたなら、ダウンスケールしてコピーを別にJPGで保存する",
-    "Use original name for output filename during batch process in extras tab": "その他タブでバッチ処理をする際、元のファイル名を出力ファイル名に使う",
-    "When using 'Save' button, only save a single selected image": "\"保存\"ボタンを使うとき、単一の選択された画像のみを保存する",
-    "Do not add watermark to images": "電子透かしを画像に追加しない",
-    "Paths for saving": "保存する場所",
-    "Output directory for images; if empty, defaults to three directories below": "画像の保存先フォルダ(下項目のデフォルト値になります)",
-    "Output directory for txt2img images": "txt2imgで作った画像の保存先フォルダ",
-    "Output directory for img2img images": "img2imgで作った画像の保存先フォルダ",
-    "Output directory for images from extras tab": "その他タブで作った画像の保存先フォルダ",
-    "Output directory for grids; if empty, defaults to two directories below": "画像の保存先フォルダ(下項目のデフォルト値になります)",
-    "Output directory for txt2img grids": "txt2imgで作ったグリッドの保存先フォルダ",
-    "Output directory for img2img grids": "img2imgで作ったグリッドの保存先フォルダ",
-    "Directory for saving images using the Save button": "保存ボタンを押したときの画像の保存先フォルダ",
-    "Saving to a directory": "フォルダについて",
-    "Save images to a subdirectory": "画像をサブフォルダに保存する",
-    "Save grids to a subdirectory": "グリッドをサブフォルダに保存する",
-    "When using \"Save\" button, save images to a subdirectory": "保存ボタンを押した時、画像をサブフォルダに保存する",
-    "Directory name pattern": "フォルダ名のパターン",
-    "Max prompt words for [prompt_words] pattern": "Max prompt words for [prompt_words] pattern",
-    "Upscaling": "アップスケール",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "ESRGANのタイルサイズ。0とするとタイルしない。",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "ESRGANのタイルの重複部分のピクセル数。少なくするとつなぎ目が見えやすくなる。",
-    "Tile size for all SwinIR.": "SwinIRのタイルサイズ",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "SwinIRのタイルの重複部分のピクセル数。少なくするとつなぎ目が見えやすくなる。",
-    "LDSR processing steps. Lower = faster": "LDSR processing steps. Lower = faster",
-    "Upscaler for img2img": "img2imgで使うアップスケーラー",
-    "Upscale latent space image when doing hires. fix": "高解像度 fix時に潜在空間(latent space)の画像をアップスケールする",
-    "Face restoration": "顔修復",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormerの重みパラメーター;0が最大で1が最小",
-    "Move face restoration model from VRAM into RAM after processing": "処理終了後、顔修復モデルをVRAMからRAMへと移動する",
-    "System": "システム設定",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "生成中のVRAM使用率の取得間隔。0にすると取得しない。",
-    "Always print all generation info to standard output": "常にすべての生成に関する情報を標準出力(stdout)に出力する",
-    "Add a second progress bar to the console that shows progress for an entire job.": "ジョブ全体の進捗をコンソールに表示する2つ目のプログレスバーを追加する",
-    "Training": "学習",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "hypernetworkの学習をするとき、VAEとCLIPをRAMへ退避する。VRAMが節約できます。",
-    "Filename word regex": "ファイル名の正規表現(学習用)",
-    "Filename join string": "ファイル名の結合子",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Number of repeats for a single input image per epoch; used only for displaying epoch number",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Save an csv containing the loss to log directory every N steps, 0 to disable",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "RAMにキャッシュするCheckpoint数",
-    "Hypernetwork strength": "Hypernetwork strength",
-    "Apply color correction to img2img results to match original colors.": "元画像に合わせてimg2imgの結果を色補正する",
-    "Save a copy of image before applying color correction to img2img results": "色補正をする前の画像も保存する",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "img2imgでスライダーで指定されたステップ数を正確に実行する（通常は、ノイズ除去を少なくするためにより少ないステップ数で実行します）。",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "より良い結果を得るために、Kサンプラーで量子化を有効にします。これにより既存のシードが変更される可能性があります。適用するには再起動が必要です。",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "強調: (text)とするとモデルはtextをより強く扱い、[text]とするとモデルはtextをより弱く扱います。",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "古い強調の実装を使う。古い生成物を再現するのに使えます。",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "K-diffusionサンプラーによるバッチ生成時に、単一画像生成時と同じ画像を生成する",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "75トークン以上を使用する場合、nトークン内の最後のカンマからパディングして一貫性を高める",
-    "Filter NSFW content": "NSFW(≒R-18)なコンテンツを検閲する",
-    "Interrogate Options": "Interrogate 設定",
-    "Interrogate: keep models in VRAM": "Interrogate: モデルをVRAMに保持する",
-    "Interrogate: use artists from artists.csv": "Interrogate: artists.csvにある芸術家などの名称を利用する",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).",
-    "Interrogate: num_beams for BLIP": "Interrogate: num_beams for BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Interrogate: minimum description length (excluding artists, etc..)",
-    "Interrogate: maximum description length": "Interrogate: maximum description length",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: maximum number of lines in text file (0 = No limit)",
-    "Interrogate: deepbooru score threshold": "Interrogate: deepbooruで拾う単語のスコア閾値",
-    "Interrogate: deepbooru sort alphabetically": "Interrogate: deepbooruで単語をアルファベット順に並べる",
-    "use spaces for tags in deepbooru": "deepbooruのタグでスペースを使う",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "deepbooruで括弧をエスケープする(\\) (強調を示す()ではなく、文字通りの()であることをモデルに示すため)",
-    "User interface": "UI設定",
-    "Show progressbar": "プログレスバーを表示",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "指定したステップ数ごとに画像の生成過程を表示する。0で無効化。",
-    "Show previews of all images generated in a batch as a grid": "Show previews of all images generated in a batch as a grid",
-    "Show grid in results for web": "WebUI上でグリッド表示",
-    "Do not show any images in results for web": "WebUI上で一切画像を表示しない",
-    "Add model hash to generation information": "モデルのハッシュ値を生成情報に追加",
-    "Add model name to generation information": "モデルの名称を生成情報に追加",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "テキストからUIに生成パラメータを読み込む場合(PNG情報または貼り付けられたテキストから)、選択されたモデル/チェックポイントは変更しない。",
-    "Font for image grids that have text": "画像グリッド内のテキストフォント",
-    "Enable full page image viewer": "フルページの画像ビューワーを有効化",
-    "Show images zoomed in by default in full page image viewer": "フルページ画像ビューアでデフォルトで画像を拡大して表示する",
-    "Show generation progress in window title.": "ウィンドウのタイトルで生成の進捗を表示",
-    "Quicksettings list": "クイック設定",
-    "Localization (requires restart)": "言語 (プログラムの再起動が必要)",
-    "ja_JP": "ja_JP",
-    "ru_RU": "ru_RU",
-    "Sampler parameters": "サンプラー parameters",
-    "Hide samplers in user interface (requires restart)": "使わないサンプリングアルゴリズムを隠す (再起動が必要)",
-    "eta (noise multiplier) for DDIM": "DDIMで用いるeta (noise multiplier)",
-    "eta (noise multiplier) for ancestral samplers": "ancestral サンプラーで用いるeta (noise multiplier)",
-    "img2img DDIM discretize": "img2img DDIM discretize",
-    "uniform": "uniform",
-    "quad": "quad",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma noise",
-    "Eta noise seed delta": "Eta noise seed delta",
-    "Images Browser": "画像閲覧",
-    "Preload images at startup": "起動時に画像を読み込んでおく",
-    "Number of pictures displayed on each page": "各ページに表示される画像の枚数",
-    "Minimum number of pages per load": "Minimum number of pages per load",
-    "Number of grids in each row": "Number of grids in each row",
-    "Request browser notifications": "ブラウザ通知の許可を要求する",
-    "Download localization template": "ローカライゼーション用のテンプレートをダウンロードする",
-    "Reload custom script bodies (No ui updates, No restart)": "カスタムスクリプトを再読み込み (UIは変更されず、再起動もしません。)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Gradioを再起動してコンポーネントをリフレッシュする (Custom Scripts, ui.py, js, cssのみ影響を受ける)",
-    "Audio": "音声",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "プロンプト (Ctrl+Enter か Alt+Enter を押して生成)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "ネガティブ プロンプト (Ctrl+Enter か Alt+Enter を押して生成)",
-    "Add a random artist to the prompt.": "芸術家などの名称をプロンプトに追加",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "プロンプトから生成パラメータを読み込むか、プロンプトが空の場合は最後の生成パラメータをユーザーインターフェースに読み込む。",
-    "Save style": "スタイルを保存する",
-    "Apply selected styles to current prompt": "現在のプロンプトに選択したスタイルを適用する",
-    "Stop processing current image and continue processing.": "現在の処理を中断し、その後の処理は続ける",
-    "Stop processing images and return any results accumulated so far.": "処理を中断し、それまでに出来た結果を表示する",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "適用するスタイル。スタイルは、ポジティブプロンプトとネガティブプロンプトの両方のコンポーネントを持ち、両方に適用される。",
-    "Do not do anything special": "特別なことをなにもしない",
-    "Which algorithm to use to produce the image": "どのアルゴリズムを使って生成するか",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - 非常に独創的で、ステップ数によって全く異なる画像が得られる、ステップ数を30～40より高く設定しても効果がない。",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Models - 描き直しには最適",
-    "Produce an image that can be tiled.": "タイルとして扱える画像を生成する",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "2ステップで、まず部分的に小さい解像度で画像を作成し、その後アップスケールすることで、構図を変えずにディテールが改善されます。",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "アルゴリズムが画像の内容をどの程度参考にするかを決定します。0 にすると何も変わりませんし、 1 にすると全く無関係な画像になります。1.0未満の値ではスライダーで指定したサンプリングステップ数よりも少ないステップ数で処理が行われます。",
-    "How many batches of images to create": "バッチ処理を何回行うか",
-    "How many image to create in a single batch": "1回のバッチ処理で何枚の画像を生成するか",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - 生成する画像がどの程度プロンプトに沿ったものになるか。 - 低い値の方がよりクリエイティブな結果を生み出します。",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "乱数発生器の出力を決定する値。同じパラメータとシードで画像を作成すれば、同じ結果が得られます。",
-    "Set seed to -1, which will cause a new random number to be used every time": "シード値を-1に設定。つまり、毎回ランダムに生成します。",
-    "Reuse seed from last generation, mostly useful if it was randomed": "前回生成時のシード値を読み出す。(ランダム生成時に便利)",
-    "Seed of a different picture to be mixed into the generation.": "生成時に混合されることになる画像のシード値",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Variationの強度。0の場合、何の効果もありません。1では、バリエーションシードで完全な画像を得ることができます（Ancestalなアルゴリズム以外では、何か(?)を得るだけです）。",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "同じシードで指定された解像度の似た画像を生成することを試みる。",
-    "This text is used to rotate the feature space of the imgs embs": "This text is used to rotate the feature space of the imgs embs",
-    "Separate values for X axis using commas.": "X軸に用いる値をカンマ(,)で区切って入力してください。",
-    "Separate values for Y axis using commas.": "Y軸に用いる値をカンマ(,)で区切って入力してください。",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "画像はフォルダ(デフォルト:log/images)に、生成パラメータはcsvファイルに書き出します。",
-    "Open images output directory": "画像の出力フォルダを開く",
-    "How much to blur the mask before processing, in pixels.": "処理前にどれだけマスクをぼかすか。px単位。",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Stable Diffusionにわたす前にマスクされたエリアに何を書き込むか",
-    "fill it with colors of the image": "元画像の色で埋める",
-    "keep whatever was there originally": "もともとあったものをそのままにする",
-    "fill it with latent space noise": "潜在空間(latent space)におけるノイズで埋める",
-    "fill it with latent space zeroes": "潜在空間(latent space)における0で埋める",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "マスクされた領域をターゲット解像度にアップスケールし、インペイントを行い、元の解像度にダウンスケールして元の画像に貼り付けます。",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "画像をターゲット解像度にリサイズします。高さと幅が一致しない場合、アスペクト比が正しくなくなります。",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "対象の解像度に画像をフィットさせます。はみ出た部分は切り取られます。",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "画像をリサイズして、ターゲット解像度の中に収まるようにします。空白部分は画像の色で埋めます。",
-    "How many times to repeat processing an image and using it as input for the next iteration": "何回画像処理を繰り返し、次の反復処理の入力として使用するか",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "ループバックモードにおいて、各ループでのノイズ除去の強度はこの値によって乗算されます。1より小さければ変化が小さくなっていって、生成される画像は1つの画像に収束します。1より大きいとどんどん変化が大きくなるので、生成される画像はよりカオスになります。",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "SDアップスケールで、どれだけタイル間の重なりを確保するか(px単位)。タイルの一部を重複させることで、1枚の画像にした時明らかな継ぎ目がなくなります。",
-    "A directory on the same machine where the server is running.": "サーバーが稼働しているのと同じマシンのあるフォルダ",
-    "Leave blank to save images to the default path.": "空欄でデフォルトの場所へ画像を保存",
-    "Input images directory": "Input images directory",
-    "Result = A * (1 - M) + B * M": "出力されるモデル = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "出力されるモデル = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "最初と最後の数字は1でなければなりません。 例:'1, 2, 1'",
-    "Path to directory with input images": "入力ファイルのあるフォルダの場所",
-    "Path to directory where to write outputs": "出力を書き込むフォルダの場所",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "以下のタグを用いてファイル名パターンを決められます: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; 空白でデフォルト設定。",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "このオプションを有効にすると、作成された画像にウォーターマークが追加されなくなります。警告:ウォーターマークを追加しない場合、非倫理的な行動とみなされる場合があります。",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "以下のタグを用いてサブフォルダのフォルダ名パターンを決められます: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; 空白でデフォルト設定",
-    "Restore low quality faces using GFPGAN neural network": "GFPGANを用いて低クオリティーな顔画像を修復",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "この正規表現を使ってファイル名から単語を抽出し、以下のオプションで結合して学習用のラベルテキストにします。ファイル名のテキストをそのまま使用する場合は、空白にしてください。",
-    "This string will be used to join split words into a single line if the option above is enabled.": "この文字列は、上記のオプションが有効な場合に、分割された単語を1行に結合するために使用されます。",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "上部のクイックアクセスバーに置く設定の設定名をカンマで区切って入力。設定名については modules/shared.py を参照してください。適用するには再起動が必要です。",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "この値が0以外の場合、シードに追加され、Etaでサンプラーを使用する際のノイズ用の乱数生成器を初期化するのに使用されます。これを利用して、さらにバリエーション豊かな画像を作成したり、他のソフトの画像に合わせたりすることができます。",
-    "NAIConvert": "NAIから変換",
-    "History": "履歴",
-    "Enable Autocomplete": "自動補完を有効化"
-}

localizations/ko_KR.json

@@ -1,592 +0,0 @@
-{
-  "×": "×",
-  "•": "•",
-  "⊞": "⊞",
-  "❮": "❮",
-  "❯": "❯",
-  "⤡": "⤡",
-  " images during ": "개의 이미지를 불러왔고, 생성 기간은 ",
-  " images in this directory. Loaded ": "개의 이미지가 이 경로에 존재합니다. ",
-  " pages": "페이지로 나뉘어 표시합니다.",
-  ", divided into ": "입니다. ",
-  ". Use Installed tab to restart.": "에 성공적으로 설치하였습니다. 설치된 확장기능 탭에서 UI를 재시작해주세요.",
-  "1st and last digit must be 1. ex:'1, 2, 1'": "1st and last digit must be 1. ex:'1, 2, 1'",
-  "[wiki]": " [위키] 참조",
-  "A directory on the same machine where the server is running.": "WebUI 서버가 돌아가고 있는 디바이스에 존재하는 디렉토리를 선택해 주세요.",
-  "A merger of the two checkpoints will be generated in your": "체크포인트들이 병합된 결과물이 당신의",
-  "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "난수 생성기의 결과물을 지정하는 값 - 동일한 설정값과 동일한 시드를 적용 시, 완전히 똑같은 결과물을 얻게 됩니다.",
-  "Action": "작업",
-  "Add a button to convert the prompts used in NovelAI for use in the WebUI. In addition, add a button that allows you to recall a previously used prompt.": "NovelAI에서 사용되는 프롬프트를 WebUI에서 사용할 수 있게 변환하는 버튼을 추가합니다. 덤으로 이전에 사용한 프롬프트를 불러오는 버튼도 추가됩니다.",
-  "Add a random artist to the prompt.": "프롬프트에 랜덤한 작가 추가",
-  "Add a second progress bar to the console that shows progress for an entire job.": "콘솔에 전체 작업의 진행도를 보여주는 2번째 프로그레스 바 추가하기",
-  "Add difference": "차이점 추가",
-  "Add extended info (seed, prompt) to filename when saving grid": "그리드 저장 시 파일명에 추가 정보(시드, 프롬프트) 기입",
-  "Add layer normalization": "레이어 정규화(normalization) 추가",
-  "Add model hash to generation information": "생성 정보에 모델 해시 추가",
-  "Add model name to generation information": "생성 정보에 모델 이름 추가",
-  "Add number to filename when saving": "이미지를 저장할 때 파일명에 숫자 추가하기",
-  "Adds a tab that lets you preview how CLIP model would tokenize your text.": "CLIP 모델이 텍스트를 어떻게 토큰화할지 미리 보여주는 탭을 추가합니다.",
-  "Adds a tab to the webui that allows the user to automatically extract keyframes from video, and manually extract 512x512 crops of those frames for use in model training.": "WebUI에 비디오로부터 자동으로 키프레임을 추출하고, 그 키프레임으로부터 모델 훈련에 사용될 512x512 이미지를 잘라낼 수 있는 탭을 추가합니다.",
-  "Aesthetic Gradients": "스타일 그라디언트",
-  "Aesthetic Image Scorer": "스타일 이미지 스코어러",
-  "Aesthetic imgs embedding": "스타일 이미지 임베딩",
-  "Aesthetic learning rate": "스타일 학습 수",
-  "Aesthetic steps": "스타일 스텝 수",
-  "Aesthetic text for imgs": "스타일 텍스트",
-  "Aesthetic weight": "스타일 가중치",
-  "Allowed categories for random artists selection when using the Roll button": "랜덤 버튼을 눌러 무작위 작가를 선택할 때 허용된 카테고리",
-  "Allows you to include various shortcodes in your prompts. You can pull text from files, set up your own variables, process text through conditional functions, and so much more - it's like wildcards on steroids.": "프롬프트에 다양한 숏코드를 추가할 수 있게 해줍니다. 파일로부터 텍스트 추출, 변수 설정, 조건 함수로 텍스트 처리 등등 - 스테로이드를 맞은 와일드카드라 할 수 있죠.",
-  "Always print all generation info to standard output": "기본 아웃풋에 모든 생성 정보 항상 출력하기",
-  "Always save all generated image grids": "생성된 이미지 그리드 항상 저장하기",
-  "Always save all generated images": "생성된 이미지 항상 저장하기",
-  "api": "",
-  "append": "뒤에 삽입",
-  "Append commas": "쉼표 삽입",
-  "Apply and restart UI": "적용 후 UI 재시작",
-  "Apply color correction to img2img results to match original colors.": "이미지→이미지 결과물이 기존 색상과 일치하도록 색상 보정 적용하기",
-  "Apply selected styles to current prompt": "현재 프롬프트에 선택된 스타일 적용",
-  "Apply settings": "설정 적용하기",
-  "Artists to study": "연구할만한 작가들",
-  "auto": "자동",
-  "Auto focal point crop": "초점 기준 크롭(자동 감지)",
-  "Autocomplete options": "자동완성 설정",
-  "Available": "지원되는 확장기능 목록",
-  "Batch count": "배치 수",
-  "Batch from Directory": "저장 경로로부터 여러장 처리",
-  "Batch img2img": "이미지→이미지 배치",
-  "Batch Process": "이미지 여러장 처리",
-  "Batch size": "배치 크기",
-  "behind": "최신 아님",
-  "Booru tag autocompletion": "Booru 태그 자동완성",
-  "BSRGAN 4x": "BSRGAN 4x",
-  "built with gradio": "gradio로 제작되었습니다",
-  "Calculates aesthetic score for generated images using CLIP+MLP Aesthetic Score Predictor based on Chad Scorer": "Chad 스코어러를 기반으로 한 CLIP+MLP 스타일 점수 예측기를 이용해 생성된 이미지의 스타일 점수를 계산합니다.",
-  "Cancel generate forever": "반복 생성 취소",
-  "cfg cnt": "CFG 변화 횟수",
-  "cfg count": "CFG 변화 횟수",
-  "CFG Scale": "CFG 스케일",
-  "cfg1 min/max": "CFG1 최소/최대",
-  "cfg2 min/max": "CFG2 최소/최대",
-  "Check for updates": "업데이트 확인",
-  "Check progress": "진행도 체크",
-  "Check progress (first)": "진행도 체크 (처음)",
-  "checkpoint": " 체크포인트 ",
-  "Checkpoint Merger": "체크포인트 병합",
-  "Checkpoint name": "체크포인트 이름",
-  "Checkpoints to cache in RAM": "RAM에 캐싱할 체크포인트 수",
-  "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - 이미지가 주어진 프롬프트를 얼마나 따를지를 정해주는 수치 - 낮은 값일수록 더 창의적인 결과물이 나옴",
-  "Click to Upload": "클릭해서 업로드하기",
-  "Clip skip": "클립 건너뛰기",
-  "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP : 텍스트 파일 최대 라인 수 (0 = 제한 없음)",
-  "CodeFormer visibility": "CodeFormer 가시성",
-  "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormer 가중치 (0 = 최대 효과, 1 = 최소 효과)",
-  "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormer 가중치 설정값 (0 = 최대 효과, 1 = 최소 효과)",
-  "Collect": "즐겨찾기",
-  "Color variation": "색깔 다양성",
-  "Combinations": "조합",
-  "Combinatorial batches": "조합 배치 수",
-  "Combinatorial generation": "조합 생성",
-  "copy": "복사",
-  "Create a grid where images will have different parameters. Use inputs below to specify which parameters will be shared by columns and rows": "서로 다른 설정값으로 생성된 이미지의 그리드를 만듭니다. 아래의 설정으로 가로/세로에 어떤 설정값을 적용할지 선택하세요.",
-  "Create a text file next to every image with generation parameters.": "생성된 이미지마다 생성 설정값을 담은 텍스트 파일 생성하기",
-  "Create aesthetic images embedding": "스타일 이미지 임베딩 생성하기",
-  "Create an embedding from one or few pictures and use it to apply their style to generated images.": "하나 혹은 그 이상의 이미지들로부터 임베딩을 생성해, 그 이미지들의 스타일을 다른 이미지 생성 시 적용할 수 있게 해줍니다.",
-  "Create debug image": "디버그 이미지 생성",
-  "Create embedding": "임베딩 생성",
-  "Create flipped copies": "좌우로 뒤집은 복사본 생성",
-  "Create hypernetwork": "하이퍼네트워크 생성",
-  "Create images embedding": "이미지 임베딩 생성하기",
-  "Crop and resize": "잘라낸 후 리사이징",
-  "Crop to fit": "잘라내서 맞추기",
-  "custom fold": "커스텀 경로",
-  "Custom Name (Optional)": "병합 모델 이름 (선택사항)",
-  "Dataset directory": "데이터셋 경로",
-  "Dataset Tag Editor": "데이터셋 태그 편집기",
-  "date": "생성 일자",
-  "DDIM": "DDIM",
-  "Decode CFG scale": "디코딩 CFG 스케일",
-  "Decode steps": "디코딩 스텝 수",
-  "Delete": "삭제",
-  "delete next": "선택한 이미지부터 시작해서 삭제할 이미지 갯수",
-  "Denoising": "디노이징",
-  "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Models - 인페이팅에 뛰어남",
-  "Denoising strength": "디노이즈 강도",
-  "Denoising strength change factor": "디노이즈 강도 변경 배수",
-  "Description": "설명",
-  "Destination directory": "결과물 저장 경로",
-  "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "알고리즘이 얼마나 원본 이미지를 반영할지를 결정하는 수치입니다. 0일 경우 아무것도 바뀌지 않고, 1일 경우 원본 이미지와 전혀 관련없는 결과물을 얻게 됩니다. 1.0 아래의 값일 경우, 설정된 샘플링 스텝 수보다 적은 스텝 수를 거치게 됩니다.",
-  "Directory for saving images using the Save button": "저장 버튼을 이용해 저장하는 이미지들의 저장 경로",
-  "Directory name pattern": "디렉토리명 패턴",
-  "directory.": "저장 경로에 저장됩니다.",
-  "Displays autocompletion hints for tags from image booru boards such as Danbooru. Uses local tag CSV files and includes a config for customization.": "Danbooru 같은 이미지 booru 보드의 태그에 대한 자동완성 힌트를 보여줍니다. 로컬 환경에 저장된 CSV 파일을 사용하고 조정 가능한 설정 파일이 포함되어 있습니다.",
-  "Do not add watermark to images": "이미지에 워터마크 추가하지 않기",
-  "Do not do anything special": "아무것도 하지 않기",
-  "Do not save grids consisting of one picture": "이미지가 1개뿐인 그리드는 저장하지 않기",
-  "Do not show any images in results for web": "웹에서 결과창에 아무 이미지도 보여주지 않기",
-  "down": "아래쪽",
-  "Download": "다운로드",
-  "Download localization template": "현지화 템플릿 다운로드",
-  "DPM adaptive": "DPM adaptive",
-  "DPM fast": "DPM fast",
-  "DPM2": "DPM2",
-  "DPM2 a": "DPM2 a",
-  "DPM2 a Karras": "DPM2 a Karras",
-  "DPM2 Karras": "DPM2 Karras",
-  "Draw legend": "범례 그리기",
-  "Draw mask": "마스크 직접 그리기",
-  "Drop File Here": "파일을 끌어 놓으세요",
-  "Drop Image Here": "이미지를 끌어 놓으세요",
-  "Dropdown": "드롭다운",
-  "Dynamic Prompts": "다이나믹 프롬프트",
-  "Embedding": "임베딩",
-  "Embedding Learning rate": "임베딩 학습률",
-  "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "강조 : (텍스트)를 이용해 모델의 텍스트에 대한 가중치를 더 강하게 주고 [텍스트]를 이용해 더 약하게 줍니다.",
-  "Enable Autocomplete": "태그 자동완성 사용",
-  "Enable full page image viewer": "전체 페이지 이미지 뷰어 활성화",
-  "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "더 예리하고 깔끔한 결과물을 위해 K 샘플러들에 양자화를 적용합니다. 존재하는 시드가 변경될 수 있습니다. 재시작이 필요합니다.",
-  "End Page": "마지막 페이지",
-  "Enter hypernetwork layer structure": "하이퍼네트워크 레이어 구조 입력",
-  "Error": "오류",
-  "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "deepbooru에서 괄호를 역슬래시(\\)로 이스케이프 처리하기(가중치 강조가 아니라 실제 괄호로 사용되게 하기 위해)",
-  "ESRGAN_4x": "ESRGAN_4x",
-  "Eta": "Eta",
-  "eta (noise multiplier) for ancestral samplers": "ancestral 샘플러를 위한 eta(노이즈 배수)값",
-  "eta (noise multiplier) for DDIM": "DDIM을 위한 eta(노이즈 배수)값",
-  "Eta noise seed delta": "Eta 노이즈 시드 변화",
-  "Euler": "Euler",
-  "Euler a": "Euler a",
-  "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - 매우 창의적, 스텝 수에 따라 완전히 다른 결과물이 나올 수 있음. 30~40보다 높은 스텝 수는 효과가 미미함",
-  "Existing Caption txt Action": "이미 존재하는 캡션 텍스트 처리",
-  "Extension": "확장기능",
-  "Extension index URL": "확장기능 목록 URL",
-  "Extensions": "확장기능",
-  "Extra": "고급",
-  "Extras": "부가기능",
-  "extras": "부가기능",
-  "extras history": "부가기능 기록",
-  "Face restoration": "얼굴 보정",
-  "Face restoration model": "얼굴 보정 모델",
-  "Fall-off exponent (lower=higher detail)": "감쇠 지수 (낮을수록 디테일이 올라감)",
-  "Favorites": "즐겨찾기",
-  "File": "파일",
-  "File format for grids": "그리드 이미지 파일 형식",
-  "File format for images": "이미지 파일 형식",
-  "File Name": "파일 이름",
-  "File with inputs": "설정값 파일",
-  "Filename join string": "파일명 병합 문자열",
-  "Filename word regex": "파일명 정규표현식",
-  "fill": "채우기",
-  "fill it with colors of the image": "이미지의 색상으로 채우기",
-  "fill it with latent space noise": "잠재 공간 노이즈로 채우기",
-  "fill it with latent space zeroes": "잠재 공간의 0값으로 채우기",
-  "Filter NSFW content": "성인 컨텐츠 필터링하기",
-  "First Page": "처음 페이지",
-  "Firstpass height": "초기 세로길이",
-  "Firstpass width": "초기 가로길이",
-  "Fixed seed": "시드 고정",
-  "Focal point edges weight": "경계면 가중치",
-  "Focal point entropy weight": "엔트로피 가중치",
-  "Focal point face weight": "얼굴 가중치",
-  "Font for image grids that have text": "텍스트가 존재하는 그리드 이미지의 폰트",
-  "for detailed explanation.": "를 참조하십시오.",
-  "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "SD 업스케일링에서 타일 간 몇 픽셀을 겹치게 할지 결정하는 설정값입니다. 타일들이 다시 한 이미지로 합쳐질 때, 눈에 띄는 이음매가 없도록 서로 겹치게 됩니다.",
-  "Generate": "생성",
-  "Generate forever": "반복 생성",
-  "Generate Info": "생성 정보",
-  "GFPGAN visibility": "GFPGAN 가시성",
-  "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "그리드 세로줄 수 : -1로 설정 시 자동 감지/0으로 설정 시 배치 크기와 동일",
-  "Height": "세로",
-  "Heun": "Heun",
-  "hide": "api 숨기기",
-  "Hide samplers in user interface (requires restart)": "사용자 인터페이스에서 숨길 샘플러 선택(재시작 필요)",
-  "Highres. fix": "고해상도 보정",
-  "History": "기록",
-  "how fast should the training go. Low values will take longer to train, high values may fail to converge (not generate accurate results) and/or may break the embedding (This has happened if you see Loss: nan in the training info textbox. If this happens, you need to manually restore your embedding from an older not-broken backup).\n\nYou can set a single numeric value, or multiple learning rates using the syntax:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nWill train with rate of 0.005 for first 100 steps, then 1e-3 until 1000 steps, then 1e-5 for all remaining steps.": "훈련이 얼마나 빨리 이루어질지 정하는 값입니다. 값이 낮을수록 훈련 시간이 길어지고, 높은 값일수록 정확한 결과를 내는 데 실패하고 임베딩을 망가뜨릴 수 있습니다(임베딩이 망가진 경우에는 훈련 정보 텍스트박스에 손실(Loss) : nan 이라고 출력되게 됩니다. 이 경우에는 망가지지 않은 이전 백업본을 불러와야 합니다).\n\n학습률은 하나의 값으로 설정할 수도 있고, 다음 문법을 사용해 여러 값을 사용할 수도 있습니다 :\n\n학습률_1:최대 스텝수_1, 학습률_2:최대 스텝수_2, ...\n\n예 : 0.005:100, 1e-3:1000, 1e-5\n\n예의 설정값은 첫 100스텝동안 0.005의 학습률로, 그 이후 1000스텝까지는 1e-3으로, 남은 스텝은 1e-5로 훈련하게 됩니다.",
-  "How many batches of images to create": "생성할 이미지 배치 수",
-  "How many image to create in a single batch": "한 배치당 이미지 수",
-  "How many times to improve the generated image iteratively; higher values take longer; very low values can produce bad results": "생성된 이미지를 향상할 횟수; 매우 낮은 값은 만족스럽지 못한 결과물을 출력할 수 있음",
-  "How many times to repeat processing an image and using it as input for the next iteration": "이미지를 생성 후 원본으로 몇 번 반복해서 사용할지 결정하는 값",
-  "How much to blur the mask before processing, in pixels.": "이미지 생성 전 마스크를 얼마나 블러처리할지 결정하는 값. 픽셀 단위",
-  "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "바리에이션을 얼마나 줄지 정하는 수치 - 0일 경우 아무것도 바뀌지 않고, 1일 경우 바리에이션 시드로부터 생성된 이미지를 얻게 됩니다. (Ancestral 샘플러 제외 - 이 경우에는 좀 다른 무언가를 얻게 됩니다)",
-  "Hypernet str.": "하이퍼네트워크 강도",
-  "Hypernetwork": "하이퍼네트워크",
-  "Hypernetwork Learning rate": "하이퍼네트워크 학습률",
-  "Hypernetwork strength": "하이퍼네트워크 강도",
-  "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "PNG 이미지가 4MB보다 크거나 가로 또는 세로길이가 4000보다 클 경우, 다운스케일 후 JPG로 복사본 저장하기",
-  "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "이 옵션이 활성화되면 생성된 이미지에 워터마크가 추가되지 않습니다. 경고 : 워터마크를 추가하지 않는다면, 비윤리적인 행동을 하는 중일지도 모릅니다.",
-  "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "이 값이 0이 아니라면, 시드에 해당 값이 더해지고, Eta가 있는 샘플러를 사용할 때 노이즈의 RNG 조정을 위해 해당 값이 사용됩니다. 이 설정으로 더 다양한 이미지를 생성하거나, 잘 알고 계시다면 특정 소프트웨어의 결과값을 재현할 수도 있습니다.",
-  "ignore": "무시",
-  "Image": "이미지",
-  "Image Browser": "이미지 브라우저",
-  "Image browser": "이미지 브라우저",
-  "Image for img2img": "Image for img2img",
-  "Image for inpainting with mask": "마스크로 인페인팅할 이미지",
-  "Image not found (may have been already moved)": "이미지를 찾을 수 없습니다 (이미 옮겨졌을 수 있음)",
-  "Images Browser": "이미지 브라우저",
-  "Images directory": "이미지 경로",
-  "Images filename pattern": "이미지 파일명 패턴",
-  "img2img": "이미지→이미지",
-  "img2img alternative test": "이미지→이미지 대체버전 테스트",
-  "img2img DDIM discretize": "이미지→이미지 DDIM 이산화",
-  "img2img history": "이미지→이미지 기록",
-  "Implements an expressive template language for random or combinatorial prompt generation along with features to support deep wildcard directory structures.": "무작위/조합 프롬프트 생성을 위한 문법과 복잡한 와일드카드 구조를 지원합니다.",
-  "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "루프백 모드에서는 매 루프마다 디노이즈 강도에 이 값이 곱해집니다. 1보다 작을 경우 다양성이 낮아져 결과 이미지들이 고정된 형태로 모일 겁니다. 1보다 클 경우 다양성이 높아져 결과 이미지들이 갈수록 혼란스러워지겠죠.",
-  "Include Separate Images": "분리된 이미지 포함하기",
-  "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "75개보다 많은 토큰을 사용시 마지막 쉼표로부터 N개의 토큰 이내에 패딩을 추가해 통일성 증가시키기",
-  "Initialization text": "초기화 텍스트",
-  "Inpaint": "인페인트",
-  "Inpaint at full resolution": "전체 해상도로 인페인트하기",
-  "Inpaint at full resolution padding, pixels": "전체 해상도로 인페인트시 패딩값(픽셀 단위)",
-  "Inpaint masked": "마스크만 처리",
-  "Inpaint not masked": "마스크 이외만 처리",
-  "Inpainting conditioning mask strength": "인페인팅 조절 마스크 강도",
-  "Input directory": "인풋 이미지 경로",
-  "Input images directory": "이미지 경로 입력",
-  "Inspiration": "\"영감\"",
-  "Install": "설치",
-  "Install from URL": "URL로부터 확장기능 설치",
-  "Installed": "설치된 확장기능",
-  "Installed into ": "확장기능을 ",
-  "Interpolation Method": "보간 방법",
-  "Interrogate\nCLIP": "CLIP\n분석",
-  "Interrogate\nDeepBooru": "DeepBooru\n분석",
-  "Interrogate Options": "분석 설정",
-  "Interrogate: deepbooru score threshold": "분석 : deepbooru 점수 임계값",
-  "Interrogate: deepbooru sort alphabetically": "분석 : deepbooru 알파벳 순서로 정렬하기",
-  "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "분석 : 결과물에 모델 태그의 랭크 포함하기 (캡션 바탕의 분석기에는 효과 없음)",
-  "Interrogate: keep models in VRAM": "분석 : VRAM에 모델 유지하기",
-  "Interrogate: maximum description length": "분석 : 설명 최대 길이",
-  "Interrogate: minimum description length (excluding artists, etc..)": "분석 : 설명 최소 길이(작가 등등..제외)",
-  "Interrogate: num_beams for BLIP": "분석 : BLIP의 num_beams값",
-  "Interrogate: use artists from artists.csv": "분석 : artists.csv의 작가들 사용하기",
-  "Interrupt": "중단",
-  "Is negative text": "네거티브 텍스트일시 체크",
-  "Iterate seed every line": "줄마다 시드 반복하기",
-  "Just resize": "리사이징",
-  "Keep -1 for seeds": "시드값 -1로 유지",
-  "keep whatever was there originally": "이미지 원본 유지",
-  "keyword": "프롬프트",
-  "Krita Plugin.": "Kirta 플러그인입니다.",
-  "Label": "라벨",
-  "Lanczos": "Lanczos",
-  "Last prompt:": "마지막 프롬프트 : ",
-  "Last saved hypernetwork:": "마지막으로 저장된 하이퍼네트워크 : ",
-  "Last saved image:": "마지막으로 저장된 이미지 : ",
-  "latent noise": "잠재 노이즈",
-  "latent nothing": "잠재 공백",
-  "latest": "최신 버전",
-  "LDSR": "LDSR",
-  "LDSR processing steps. Lower = faster": "LDSR 스텝 수. 낮은 값 = 빠른 속도",
-  "leakyrelu": "leakyrelu",
-  "Leave blank to save images to the default path.": "기존 저장 경로에 이미지들을 저장하려면 비워두세요.",
-  "Leave empty for auto": "자동 설정하려면 비워두십시오",
-  "left": "왼쪽",
-  "Lets you edit captions in training datasets.": "훈련에 사용되는 데이터셋의 캡션을 수정할 수 있게 해줍니다.",
-  "linear": "linear",
-  "List of prompt inputs": "프롬프트 입력 리스트",
-  "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "설정 탭이 아니라 상단의 빠른 설정 바에 위치시킬 설정 이름을 쉼표로 분리해서 입력하십시오. 설정 이름은 modules/shared.py에서 찾을 수 있습니다. 재시작이 필요합니다.",
-  "LMS": "LMS",
-  "LMS Karras": "LMS Karras",
-  "Load": "불러오기",
-  "Load from:": "URL로부터 불러오기",
-  "Loading...": "로딩 중...",
-  "Local directory name": "로컬 경로 이름",
-  "Localization (requires restart)": "현지화 (재시작 필요)",
-  "Log directory": "로그 경로",
-  "Loopback": "루프백",
-  "Loops": "루프 수",
-  "Loss:": "손실(Loss) : ",
-  "Magic prompt": "매직 프롬프트",
-  "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "동일한 시드 값으로 생성되었을 이미지를 주어진 해상도로 최대한 유사하게 재현합니다.",
-  "Make K-diffusion samplers produce same images in a batch as when making a single image": "K-diffusion 샘플러들이 단일 이미지를 생성하는 것처럼 배치에서도 동일한 이미지를 생성하게 하기",
-  "Make Zip when Save?": "저장 시 Zip 생성하기",
-  "Mask": "마스크",
-  "Mask blur": "마스크 블러",
-  "Mask mode": "마스크 모드",
-  "Masked content": "마스크된 부분",
-  "Masking mode": "마스킹 모드",
-  "Max prompt words for [prompt_words] pattern": "[prompt_words] 패턴의 최대 프롬프트 단어 수",
-  "Max steps": "최대 스텝 수",
-  "Minimum number of pages per load": "한번 불러올 때마다 불러올 최소 페이지 수",
-  "Modules": "모듈",
-  "Move face restoration model from VRAM into RAM after processing": "처리가 완료되면 얼굴 보정 모델을 VRAM에서 RAM으로 옮기기",
-  "Move to favorites": "즐겨찾기로 옮기기",
-  "Move VAE and CLIP to RAM when training if possible. Saves VRAM.": "훈련 진행 시 가능하면 VAE와 CLIP을 RAM으로 옮기기. VRAM이 절약됩니다.",
-  "Moved to favorites": "즐겨찾기로 옮겨짐",
-  "Multiplier (M) - set to 0 to get model A": "배율 (M) - 0으로 적용하면 모델 A를 얻게 됩니다",
-  "Name": "이름",
-  "Negative prompt": "네거티브 프롬프트",
-  "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "네거티브 프롬프트(Prompt) 입력(Ctrl+Enter나 Alt+Enter로 생성 시작)",
-  "Next batch": "다음 묶음",
-  "Next Page": "다음 페이지",
-  "None": "없음",
-  "Nothing": "없음",
-  "Nothing found in the image.": "Nothing found in the image.",
-  "novelai-2-local-prompt": "NovelAI 프롬프트 변환기",
-  "Number of columns on the page": "각 페이지마다 표시할 가로줄 수",
-  "Number of grids in each row": "각 세로줄마다 표시될 그리드 수",
-  "number of images to delete consecutively next": "연속적으로 삭제할 이미지 수",
-  "Number of pictures displayed on each page": "각 페이지에 표시될 이미지 수",
-  "Number of repeats for a single input image per epoch; used only for displaying epoch number": "세대(Epoch)당 단일 인풋 이미지의 반복 횟수 - 세대(Epoch) 숫자를 표시하는 데에만 사용됩니다. ",
-  "Number of rows on the page": "각 페이지마다 표시할 세로줄 수",
-  "Number of vectors per token": "토큰별 벡터 수",
-  "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "인페인팅 모델에만 적용됩니다. 인페인팅과 이미지→이미지에서 원본 이미지를 얼마나 마스킹 처리할지 결정하는 값입니다. 1.0은 완전히 마스킹함(기본 설정)을 의미하고, 0.0은 완전히 언마스킹된 이미지를 의미합니다. 낮은 값일수록 이미지의 전체적인 구성을 유지하는 데에 도움되겠지만, 변화량이 많을수록 불안정해집니다.",
-  "Open for Clip Aesthetic!": "클립 스타일 기능을 활성화하려면 클릭!",
-  "Open images output directory": "이미지 저장 경로 열기",
-  "Open output directory": "저장 경로 열기",
-  "or": "또는",
-  "original": "원본 유지",
-  "Original negative prompt": "기존 네거티브 프롬프트",
-  "Original prompt": "기존 프롬프트",
-  "Others": "기타",
-  "Outpainting direction": "아웃페인팅 방향",
-  "Outpainting mk2": "아웃페인팅 마크 2",
-  "Output directory": "이미지 저장 경로",
-  "Output directory for grids; if empty, defaults to two directories below": "그리드 이미지 저장 경로 - 비워둘 시 하단의 2가지 기본 경로로 설정됨",
-  "Output directory for images from extras tab": "부가기능 탭 저장 경로",
-  "Output directory for images; if empty, defaults to three directories below": "이미지 저장 경로 - 비워둘 시 하단의 3가지 기본 경로로 설정됨",
-  "Output directory for img2img grids": "이미지→이미지 그리드 저장 경로",
-  "Output directory for img2img images": "이미지→이미지 저장 경로",
-  "Output directory for txt2img grids": "텍스트→이미지 그리드 저장 경로",
-  "Output directory for txt2img images": "텍스트→이미지 저장 경로",
-  "Override `Denoising strength` to 1?": "디노이즈 강도를 1로 적용할까요?",
-  "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "프롬프트 값을 기존 프롬프트와 동일하게 적용할까요?(네거티브 프롬프트 포함)",
-  "Override `Sampling method` to Euler?(this method is built for it)": "샘플링 방법을 Euler로 적용할까요?(이 기능은 해당 샘플러를 위해 만들어져 있습니다)",
-  "Override `Sampling Steps` to the same value as `Decode steps`?": "샘플링 스텝 수를 디코딩 스텝 수와 동일하게 적용할까요?",
-  "Overwrite Old Embedding": "기존 임베딩 덮어쓰기",
-  "Overwrite Old Hypernetwork": "기존 하이퍼네트워크 덮어쓰기",
-  "Page Index": "페이지 인덱스",
-  "parameters": "설정값",
-  "path name": "경로 이름",
-  "Path to directory where to write outputs": "결과물을 출력할 경로",
-  "Path to directory with input images": "인풋 이미지가 있는 경로",
-  "Paths for saving": "저장 경로",
-  "Pixels to expand": "확장할 픽셀 수",
-  "PLMS": "PLMS",
-  "PNG Info": "PNG 정보",
-  "Poor man's outpainting": "가난뱅이의 아웃페인팅",
-  "Preload images at startup": "WebUI 가동 시 이미지 프리로드하기",
-  "Preparing dataset from": "준비된 데이터셋 경로 : ",
-  "prepend": "앞에 삽입",
-  "Preprocess": "전처리",
-  "Preprocess images": "이미지 전처리",
-  "Prev batch": "이전 묶음",
-  "Prev Page": "이전 페이지",
-  "Prevent empty spots in grid (when set to autodetect)": "(자동 감지 사용시)그리드에 빈칸이 생기는 것 방지하기",
-  "Primary model (A)": "주 모델 (A)",
-  "Process an image, use it as an input, repeat.": "이미지를 생성하고, 생성한 이미지를 다시 원본으로 사용하는 과정을 반복합니다.",
-  "Process images in a directory on the same machine where the server is running.": "WebUI 서버가 돌아가고 있는 디바이스에 존재하는 디렉토리의 이미지들을 처리합니다.",
-  "Produce an image that can be tiled.": "타일링 가능한 이미지를 생성합니다.",
-  "Prompt": "프롬프트",
-  "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "프롬프트(Prompt) 입력(Ctrl+Enter나 Alt+Enter로 생성 시작)",
-  "Prompt matrix": "프롬프트 매트릭스",
-  "Prompt order": "프롬프트 순서",
-  "Prompt S/R": "프롬프트 스타일 변경",
-  "Prompt template file": "프롬프트 템플릿 파일 경로",
-  "Prompts": "프롬프트",
-  "Prompts from file or textbox": "파일이나 텍스트박스로부터 프롬프트 불러오기",
-  "Provides an interface to browse created images in the web browser.": "생성된 이미지를 브라우저 내에서 볼 수 있는 인터페이스를 추가합니다.",
-  "Put variable parts at start of prompt": "변경되는 프롬프트를 앞에 위치시키기",
-  "quad": "quad",
-  "Quality for saved jpeg images": "저장된 jpeg 이미지들의 품질",
-  "Quicksettings list": "빠른 설정 리스트",
-  "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-  "Random": "랜덤",
-  "Random grid": "랜덤 그리드",
-  "Randomly display the pictures of the artist's or artistic genres typical style, more pictures of this artist or genre is displayed after selecting. So you don't have to worry about how hard it is to choose the right style of art when you create.": "특정 작가 또는 스타일의 이미지들 중 하나를 무작위로 보여줍니다. 선택 후 선택한 작가 또는 스타일의 이미지들이 더 나타나게 됩니다. 고르기 어려워도 걱정하실 필요 없어요!",
-  "Randomness": "랜덤성",
-  "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "클립보드에 복사된 정보로부터 설정값 읽어오기/프롬프트창이 비어있을경우 제일 최근 설정값 불러오기",
-  "Read parameters (prompt, etc...) from txt2img tab when making previews": "프리뷰 이미지 생성 시 텍스트→이미지 탭에서 설정값(프롬프트 등) 읽어오기",
-  "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "추천 설정값 - 샘플링 스텝 수 : 80-100 , 샘플러 : Euler a, 디노이즈 강도 : 0.8",
-  "Reload custom script bodies (No ui updates, No restart)": "커스텀 스크립트 리로드하기(UI 업데이트 없음, 재시작 없음)",
-  "Reloading...": "재시작 중...",
-  "relu": "relu",
-  "Renew Page": "Renew Page",
-  "Request browser notifications": "브라우저 알림 권한 요청",
-  "Resize": "리사이징 배수",
-  "Resize and fill": "리사이징 후 채우기",
-  "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "설정된 해상도로 이미지 리사이징을 진행합니다. 원본과 가로/세로 길이가 일치하지 않을 경우, 부정확한 화면비의 이미지를 얻게 됩니다.",
-  "Resize mode": "리사이징 모드",
-  "Resize seed from height": "시드 리사이징 가로길이",
-  "Resize seed from width": "시드 리사이징 세로길이",
-  "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "이미지 전체가 설정된 해상도 내부에 들어가게 리사이징을 진행합니다. 빈 공간은 이미지의 색상으로 채웁니다.",
-  "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "설정된 해상도 전체가 이미지로 가득차게 리사이징을 진행합니다. 튀어나오는 부분은 잘라냅니다.",
-  "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Gradio를 재시작하고 컴포넌트 새로고침하기 (커스텀 스크립트, ui.py, js, css만 해당됨)",
-  "Restore faces": "얼굴 보정",
-  "Restore low quality faces using GFPGAN neural network": "GFPGAN 신경망을 이용해 저품질의 얼굴을 보정합니다.",
-  "Result = A * (1 - M) + B * M": "결과물 = A * (1 - M) + B * M",
-  "Result = A + (B - C) * M": "결과물 = A + (B - C) * M",
-  "Reuse seed from last generation, mostly useful if it was randomed": "이전 생성에서 사용된 시드를 불러옵니다. 랜덤하게 생성했을 시 도움됨",
-  "right": "오른쪽",
-  "Run": "가동",
-  "Sample extension. Allows you to use __name__ syntax in your prompt to get a random line from a file named name.txt in the wildcards directory. Also see Dynamic Prompts for similar functionality.": "샘플 확장기능입니다. __이름__형식의 문법을 사용해 와일드카드 경로 내의 이름.txt파일로부터 무작위 프롬프트를 적용할 수 있게 해줍니다. 유사한 확장기능으로 다이나믹 프롬프트가 있습니다.",
-  "Sampler": "샘플러",
-  "Sampler parameters": "샘플러 설정값",
-  "Sampling method": "샘플링 방법",
-  "Sampling Steps": "샘플링 스텝 수",
-  "Save": "저장",
-  "Save a copy of embedding to log directory every N steps, 0 to disable": "N스텝마다 로그 경로에 임베딩을 저장합니다, 비활성화하려면 0으로 설정하십시오.",
-  "Save a copy of image before applying color correction to img2img results": "이미지→이미지 결과물에 색상 보정을 진행하기 전 이미지의 복사본을 저장하기",
-  "Save a copy of image before applying highres fix.": "고해상도 보정을 진행하기 전 이미지의 복사본을 저장하기",
-  "Save a copy of image before doing face restoration.": "얼굴 보정을 진행하기 전 이미지의 복사본을 저장하기",
-  "Save an csv containing the loss to log directory every N steps, 0 to disable": "N스텝마다 로그 경로에 손실(Loss)을 포함하는 csv 파일을 저장합니다, 비활성화하려면 0으로 설정하십시오.",
-  "Save an image to log directory every N steps, 0 to disable": "N스텝마다 로그 경로에 이미지를 저장합니다, 비활성화하려면 0으로 설정하십시오.",
-  "Save as float16": "float16으로 저장",
-  "Save grids to a subdirectory": "그리드 이미지를 하위 디렉토리에 저장하기",
-  "Save images to a subdirectory": "이미지를 하위 디렉토리에 저장하기",
-  "Save images with embedding in PNG chunks": "PNG 청크로 이미지에 임베딩을 포함시켜 저장",
-  "Save style": "스타일 저장",
-  "Save text information about generation parameters as chunks to png files": "이미지 생성 설정값을 PNG 청크에 텍스트로 저장",
-  "Saves Optimizer state as separate *.optim file. Training can be resumed with HN itself and matching optim file.": "옵티마이저 상태를 별개의 *.optim 파일로 저장하기. 하이퍼네트워크 파일과 일치하는 optim 파일로부터 훈련을 재개할 수 있습니다.",
-  "Saving images/grids": "이미지/그리드 저장",
-  "Saving to a directory": "디렉토리에 저장",
-  "Scale by": "스케일링 배수 지정",
-  "Scale to": "스케일링 사이즈 지정",
-  "Script": "스크립트",
-  "ScuNET GAN": "ScuNET GAN",
-  "ScuNET PSNR": "ScuNET PSNR",
-  "SD upscale": "SD 업스케일링",
-  "Secondary model (B)": "2차 모델 (B)",
-  "See": "자세한 설명은",
-  "Seed": "시드",
-  "Seed of a different picture to be mixed into the generation.": "결과물에 섞일 다른 그림의 시드",
-  "Select activation function of hypernetwork. Recommended : Swish / Linear(none)": "하이퍼네트워크 활성화 함수 선택 - 추천 : Swish / Linear(None)",
-  "Select Layer weights initialization. Recommended: Kaiming for relu-like, Xavier for sigmoid-like, Normal otherwise": "레이어 가중치 초기화 방식 선택 - relu류 : Kaiming 추천, sigmoid류 : Xavier 추천, 그 외 : Normal",
-  "Select which Real-ESRGAN models to show in the web UI. (Requires restart)": "WebUI에 표시할 Real-ESRGAN 모델을 선택하십시오. (재시작 필요)",
-  "Send seed when sending prompt or image to other interface": "다른 화면으로 프롬프트나 이미지를 보낼 때 시드도 함께 보내기",
-  "Send to extras": "부가기능으로 전송",
-  "Send to img2img": "이미지→이미지로 전송",
-  "Send to inpaint": "인페인트로 전송",
-  "Send to txt2img": "텍스트→이미지로 전송",
-  "Separate prompts into parts using vertical pipe character (|) and the script will create a picture for every combination of them (except for the first part, which will be present in all combinations)": "(|)를 이용해 프롬프트를 분리할 시 첫 프롬프트를 제외하고 모든 프롬프트의 조합마다 이미지를 생성합니다. 첫 프롬프트는 모든 조합에 포함되게 됩니다.",
-  "Separate values for X axis using commas.": "쉼표로 X축에 적용할 값 분리",
-  "Separate values for Y axis using commas.": "쉼표로 Y축에 적용할 값 분리",
-  "Set seed to -1, which will cause a new random number to be used every time": "시드를 -1로 적용 - 매번 랜덤한 시드가 적용되게 됩니다.",
-  "set_index": "set_index",
-  "Settings": "설정",
-  "should be 2 or lower.": "이 2 이하여야 합니다.",
-  "Show generation progress in window title.": "창 타이틀에 생성 진행도 보여주기",
-  "Show grid in results for web": "웹에서 결과창에 그리드 보여주기",
-  "Show image creation progress every N sampling steps. Set to 0 to disable. Set to -1 to show after completion of batch.": "N번째 샘플링 스텝마다 이미지 생성 과정 보이기 - 비활성화하려면 0으로 설정, 배치 생성 완료시 보이려면 -1로 설정",
-  "Show images zoomed in by default in full page image viewer": "전체 페이지 이미지 뷰어에서 기본값으로 이미지 확대해서 보여주기",
-  "Show previews of all images generated in a batch as a grid": "배치에서 생성된 모든 이미지의 미리보기를 그리드 형식으로 보여주기",
-  "Show progressbar": "프로그레스 바 보이기",
-  "Show result images": "이미지 결과 보이기",
-  "Show Textbox": "텍스트박스 보이기",
-  "Shows a gallery of generated pictures by artists separated into categories.": "생성된 이미지들을 작가별로 분류해 보여줍니다. 원본 - https://artiststostudy.pages.dev",
-  "Sigma adjustment for finding noise for image": "이미지 노이즈를 찾기 위해 시그마 조정",
-  "Sigma Churn": "시그마 섞기",
-  "sigma churn": "시그마 섞기",
-  "Sigma max": "시그마 최댓값",
-  "Sigma min": "시그마 최솟값",
-  "Sigma noise": "시그마 노이즈",
-  "sigma noise": "시그마 노이즈",
-  "sigma tmin": "시그마 tmin",
-  "Single Image": "단일 이미지",
-  "Skip": "건너뛰기",
-  "Slerp angle": "구면 선형 보간 각도",
-  "Slerp interpolation": "구면 선형 보간",
-  "sort by": "정렬 기준",
-  "Source": "원본",
-  "Source directory": "원본 경로",
-  "Split image overlap ratio": "이미지 분할 겹침 비율",
-  "Split image threshold": "이미지 분할 임계값",
-  "Split oversized images": "사이즈가 큰 이미지 분할하기",
-  "Stable Diffusion": "Stable Diffusion",
-  "Stable Diffusion checkpoint": "Stable Diffusion 체크포인트",
-  "step cnt": "스텝 변화 횟수",
-  "step count": "스텝 변화 횟수",
-  "step1 min/max": "스텝1 최소/최대",
-  "step2 min/max": "스텝2 최소/최대",
-  "Step:": "Step:",
-  "Steps": "스텝 수",
-  "Stop At last layers of CLIP model": "CLIP 모델의 n번째 레이어에서 멈추기",
-  "Stop processing current image and continue processing.": "현재 진행중인 이미지 생성을 중단하고 작업을 계속하기",
-  "Stop processing images and return any results accumulated so far.": "이미지 생성을 중단하고 지금까지 진행된 결과물 출력",
-  "Style 1": "스타일 1",
-  "Style 2": "스타일 2",
-  "Style to apply; styles have components for both positive and negative prompts and apply to both": "적용할 스타일 - 스타일은 긍정/부정 프롬프트 모두에 대한 설정값을 가지고 있고 양쪽 모두에 적용 가능합니다.",
-  "SwinIR 4x": "SwinIR 4x",
-  "Sys VRAM:": "시스템 VRAM : ",
-  "System": "시스템",
-  "Tertiary model (C)": "3차 모델 (C)",
-  "Textbox": "텍스트박스",
-  "The official port of Deforum, an extensive script for 2D and 3D animations, supporting keyframable sequences, dynamic math parameters (even inside the prompts), dynamic masking, depth estimation and warping.": "Deforum의 공식 포팅 버전입니다. 2D와 3D 애니메이션, 키프레임 시퀀스, 수학적 매개변수, 다이나믹 마스킹 등을 지원합니다.",
-  "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "이 정규표현식은 파일명으로부터 단어를 추출하는 데 사용됩니다. 추출된 단어들은 하단의 설정을 이용해 라벨 텍스트로 변환되어 훈련에 사용됩니다. 파일명 텍스트를 유지하려면 비워두십시오.",
-  "This string will be used to join split words into a single line if the option above is enabled.": "이 문자열은 상단 설정이 활성화되어있을 때 분리된 단어들을 한 줄로 합치는 데 사용됩니다.",
-  "This text is used to rotate the feature space of the imgs embs": "이 텍스트는 이미지 임베딩의 특징 공간을 회전하는 데 사용됩니다.",
-  "Tile overlap": "타일 겹침",
-  "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "ESRGAN 업스케일러들의 타일 중첩 수치, 픽셀 단위. 낮은 값 = 눈에 띄는 이음매.",
-  "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "SwinIR의 타일 중첩 수치, 픽셀 단위. 낮은 값 = 눈에 띄는 이음매.",
-  "Tile size for all SwinIR.": "SwinIR의 타일 사이즈.",
-  "Tile size for ESRGAN upscalers. 0 = no tiling.": "ESRGAN 업스케일러들의 타일 사이즈. 0 = 타일링 없음.",
-  "Tiling": "타일링",
-  "Time taken:": "소요 시간 : ",
-  "tokenizer": "토크나이저",
-  "Torch active/reserved:": "활성화/예약된 Torch 양 : ",
-  "Torch active: Peak amount of VRAM used by Torch during generation, excluding cached data.\nTorch reserved: Peak amount of VRAM allocated by Torch, including all active and cached data.\nSys VRAM: Peak amount of VRAM allocation across all applications / total GPU VRAM (peak utilization%).": "활성화된 Torch : 생성 도중 캐시된 데이터를 포함해 사용된 VRAM의 최대량\n예약된 Torch : 활성화되고 캐시된 모든 데이터를 포함해 Torch에게 할당된 VRAM의 최대량\n시스템 VRAM : 모든 어플리케이션에 할당된 VRAM 최대량 / 총 GPU VRAM (최고 이용도%)",
-  "Train": "훈련",
-  "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "임베딩이나 하이퍼네트워크를 훈련시킵니다. 1:1 비율의 이미지가 있는 경로를 지정해야 합니다.",
-  "Train Embedding": "임베딩 훈련",
-  "Train Hypernetwork": "하이퍼네트워크 훈련",
-  "Training": "훈련",
-  "training-picker": "훈련용 선택기",
-  "txt2img": "텍스트→이미지",
-  "txt2img history": "텍스트→이미지 기록",
-  "uniform": "uniform",
-  "unknown": "알수 없음",
-  "up": "위쪽",
-  "Update": "업데이트",
-  "Upload mask": "마스크 업로드하기",
-  "Upload prompt inputs": "입력할 프롬프트를 업로드하십시오",
-  "Upscale Before Restoring Faces": "얼굴 보정을 진행하기 전에 업스케일링 먼저 진행하기",
-  "Upscale latent space image when doing hires. fix": "고해상도 보정 사용시 잠재 공간 이미지 업스케일하기",
-  "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "마스크된 부분을 설정된 해상도로 업스케일하고, 인페인팅을 진행한 뒤, 다시 다운스케일 후 원본 이미지에 붙여넣습니다.",
-  "Upscaler": "업스케일러",
-  "Upscaler 1": "업스케일러 1",
-  "Upscaler 2": "업스케일러 2",
-  "Upscaler 2 visibility": "업스케일러 2 가시성",
-  "Upscaler for img2img": "이미지→이미지 업스케일러",
-  "Upscaling": "업스케일링",
-  "URL for extension's git repository": "확장기능의 git 레포 URL",
-  "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "저해상도 이미지를 1차적으로 생성 후 업스케일을 진행하여, 이미지의 전체적인 구성을 바꾸지 않고 세부적인 디테일을 향상시킵니다.",
-  "Use an empty output directory to save pictures normally instead of writing to the output directory.": "저장 경로를 비워두면 기본 저장 폴더에 이미지들이 저장됩니다.",
-  "Use BLIP for caption": "캡션에 BLIP 사용",
-  "Use checkbox to enable the extension; it will be enabled or disabled when you click apply button": "체크박스를 이용해 적용할 확장기능을 선택하세요. 변경사항은 적용 후 UI 재시작 버튼을 눌러야 적용됩니다.",
-  "Use checkbox to mark the extension for update; it will be updated when you click apply button": "체크박스를 이용해 업데이트할 확장기능을 선택하세요. 업데이트는 적용 후 UI 재시작 버튼을 눌러야 적용됩니다.",
-  "Use cross attention optimizations while training": "훈련 진행 시 크로스 어텐션 최적화 사용",
-  "Use deepbooru for caption": "캡션에 deepbooru 사용",
-  "Use dropout": "드롭아웃 사용",
-  "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "다음 태그들을 사용해 이미지 파일명 형식을 결정하세요 : [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]. 비워두면 기본값으로 설정됩니다.",
-  "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "다음 태그들을 사용해 이미지와 그리드의 하위 디렉토리명의 형식을 결정하세요 : [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]. 비워두면 기본값으로 설정됩니다.",
-  "Use old emphasis implementation. Can be useful to reproduce old seeds.": "옛 방식의 강조 구현을 사용합니다. 옛 시드를 재현하는 데 효과적일 수 있습니다.",
-  "Use original name for output filename during batch process in extras tab": "부가기능 탭에서 이미지를 여러장 처리 시 결과물 파일명에 기존 파일명 사용하기",
-  "Use same random seed for all lines": "모든 줄에 동일한 시드 사용",
-  "Use same seed for each image": "각 이미지에 동일한 시드 사용",
-  "use spaces for tags in deepbooru": "deepbooru에서 태그에 공백 사용",
-  "User interface": "사용자 인터페이스",
-  "Var. seed": "바리에이션 시드",
-  "Var. strength": "바리에이션 강도",
-  "Variation seed": "바리에이션 시드",
-  "Variation strength": "바리에이션 강도",
-  "view": "api 보이기",
-  "VRAM usage polls per second during generation. Set to 0 to disable.": "생성 도중 초당 VRAM 사용량 폴링 수. 비활성화하려면 0으로 설정하십시오.",
-  "Weighted sum": "가중 합",
-  "What to put inside the masked area before processing it with Stable Diffusion.": "Stable Diffusion으로 이미지를 생성하기 전 마스크된 부분에 무엇을 채울지 결정하는 설정값",
-  "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "PNG 정보나 붙여넣은 텍스트로부터 생성 설정값을 읽어올 때, 선택된 모델/체크포인트는 변경하지 않기.",
-  "When using \"Save\" button, save images to a subdirectory": "저장 버튼 사용시, 이미지를 하위 디렉토리에 저장하기",
-  "When using 'Save' button, only save a single selected image": "저장 버튼 사용시, 선택된 이미지 1개만 저장하기",
-  "Which algorithm to use to produce the image": "이미지를 생성할 때 사용할 알고리즘",
-  "Width": "가로",
-  "wiki": " 위키",
-  "Wildcards": "와일드카드",
-  "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "이미지를 설정된 사이즈의 2배로 업스케일합니다. 상단의 가로와 세로 슬라이더를 이용해 타일 사이즈를 지정하세요.",
-  "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "이미지→이미지 진행 시, 슬라이더로 설정한 스텝 수를 정확히 실행하기 (일반적으로 디노이즈 강도가 낮을수록 실제 설정된 스텝 수보다 적게 진행됨)",
-  "Write image to a directory (default - log/images) and generation parameters into csv file.": "이미지를 경로에 저장하고, 설정값들을 csv 파일로 저장합니다. (기본 경로 - log/images)",
-  "X type": "X축",
-  "X values": "X 설정값",
-  "x/y change": "X/Y축 변경",
-  "X/Y plot": "X/Y 플롯",
-  "Y type": "Y축",
-  "Y values": "Y 설정값"
-}

localizations/pt_BR.json

@@ -1,485 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "Carregando...",
-    "view": "ver",
-    "api": "api",
-    "•": "•",
-    "built with gradio": "criado com gradio",
-    "Stable Diffusion checkpoint": "Stable Diffusion checkpoint",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "Extras",
-    "PNG Info": "Informações de PNG",
-    "Checkpoint Merger": "Fusão de Checkpoint",
-    "Train": "Treinar",
-    "Settings": "Configurações",
-    "Extensions": "Extensões",
-    "Prompt": "Prompt",
-    "Negative prompt": "Prompt negativo",
-    "Run": "Executar",
-    "Skip": "Pular",
-    "Interrupt": "Interromper",
-    "Generate": "Gerar",
-    "Style 1": "Estilo 1",
-    "Style 2": "Estilo 2",
-    "Label": "Rótulo",
-    "File": "Arquivo",
-    "Drop File Here": "Solte Aqui o Arquivo",
-    "-": "-",
-    "or": "ou",
-    "Click to Upload": "Clique para Carregar um Arquivo",
-    "Image": "Imagem",
-    "Check progress": "Checar progresso",
-    "Check progress (first)": "Checar progresso (primeiro)",
-    "Sampling Steps": "Passos de Amostragem",
-    "Sampling method": "Método de amostragem",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Largura",
-    "Height": "Altura",
-    "Restore faces": "Restaurar rostos",
-    "Tiling": "Ladrilhar",
-    "Highres. fix": "Ajuste de alta resolução",
-    "Firstpass width": "Primeira Passagem da largura",
-    "Firstpass height": "Primeira Passagem da altura",
-    "Denoising strength": "Força do denoise",
-    "Batch count": "Quantidade por lote",
-    "Batch size": "Quantidade de lotes",
-    "CFG Scale": "Escala CFG",
-    "Seed": "Seed",
-    "Extra": "Extra",
-    "Variation seed": "Variação de seed",
-    "Variation strength": "Força da variação",
-    "Resize seed from width": "Redimensionar a seed a partir da largura",
-    "Resize seed from height": "Redimensionar a seed a partir da altura",
-    "Script": "Script",
-    "None": "Nenhum",
-    "Prompt matrix": "Matriz de prompt",
-    "Prompts from file or textbox": "Prompts a partir de arquivo ou caixa de texto",
-    "X/Y plot": "X/Y plot",
-    "Put variable parts at start of prompt": "Coloca partes variáveis no começo do prompt",
-    "Iterate seed every line": "Iterar seed a cada linha",
-    "List of prompt inputs": "Lista de entrada de texto para prompt",
-    "Upload prompt inputs": "Carregar entrada de texto para prompt",
-    "X type": "Tipo do X",
-    "Nothing": "Nenhum",
-    "Var. seed": "Var. seed",
-    "Var. strength": "Var. da força",
-    "Steps": "Passos",
-    "Prompt S/R": "Prompt S/R",
-    "Prompt order": "Ordem de Prompt",
-    "Sampler": "Sampler",
-    "Checkpoint name": "Nome do Checkpoint",
-    "Hypernetwork": "Hypernetwork",
-    "Hypernet str.": "Força da Hypernet",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma min",
-    "Sigma max": "Sigma max",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Tempo estimado",
-    "Clip skip": "Pular Clip",
-    "Denoising": "Denoising",
-    "Cond. Image Mask Weight": "Peso da Máscara Condicional de Imagem",
-    "X values": "Valores de X",
-    "Y type": "Tipo de Y",
-    "Y values": "Valores de Y",
-    "Draw legend": "Desenhar a legenda",
-    "Include Separate Images": "Incluir Imagens Separadas",
-    "Keep -1 for seeds": "Manter em -1 para seeds",
-    "Save": "Salvar",
-    "Send to img2img": "Mandar para img2img",
-    "Send to inpaint": "Mandar para inpaint",
-    "Send to extras": "Mandar para extras",
-    "Make Zip when Save?": "Criar um Zip quando salvar?",
-    "Textbox": "Caixa de texto",
-    "Interrogate\nCLIP": "Interrogatório\nCLIP",
-    "Inpaint": "Inpaint",
-    "Batch img2img": "Lote img2img",
-    "Image for img2img": "Imagem para img2img",
-    "Drop Image Here": "Solte a imagem aqui",
-    "Image for inpainting with mask": "Imagem para inpainting com máscara",
-    "Mask": "Máscara",
-    "Mask blur": "Desfoque da máscara",
-    "Mask mode": "Modo de máscara",
-    "Draw mask": "Desenhar máscara",
-    "Upload mask": "Carregar máscara",
-    "Masking mode": "Modo de máscara",
-    "Inpaint masked": "Inpaint o que está dentro da máscara",
-    "Inpaint not masked": "Inpaint o que está fora da máscara",
-    "Masked content": "Conteúdo mascarado",
-    "fill": "preencher",
-    "original": "original",
-    "latent noise": "latent noise",
-    "latent nothing": "latent nothing",
-    "Inpaint at full resolution": "Inpaint em resolução total",
-    "Inpaint at full resolution padding, pixels": "Inpaint de preenchimento em resolução total, pixels",
-    "Process images in a directory on the same machine where the server is running.": "Processar imagens no diretório da mesma maquina onde o servidor está rodando.",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Usar um diretório vazio para salvar imagens, ao invés de salvá-las no diretório output.",
-    "Input directory": "Diretório de entrada",
-    "Output directory": "Diretório de saída",
-    "Resize mode": "Modo de redimensionamento",
-    "Just resize": "Apenas redimensionar",
-    "Crop and resize": "Cortar e redimensionar",
-    "Resize and fill": "Redimensionar e preencher",
-    "img2img alternative test": "Teste alternativo de img2img",
-    "Loopback": "Loopback",
-    "Outpainting mk2": "Outpainting mk2",
-    "Poor man's outpainting": "Poor man`s outpainting",
-    "SD upscale": "Ampliamento SD",
-    "should be 2 or lower.": "deve ser 2 ou menos.",
-    "Override `Sampling method` to Euler?(this method is built for it)": "Substituir `Método de amostragem` por Euler? (este método foi feito para isso)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "Substituir `prompt` para o mesmo valor que o `prompt original`? (também para o `prompt negativo`)",
-    "Original prompt": "Prompt original",
-    "Original negative prompt": "Prompt negativo original",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "Substituir `Passos de Amostragem` para o mesmo valor que `Decodificar Passos`?",
-    "Decode steps": "Decode steps",
-    "Override `Denoising strength` to 1?": "Substituir `Quantidade do Denoise` para 1?",
-    "Decode CFG scale": "Decodificar escala CFG",
-    "Randomness": "Aleatoriedade",
-    "Sigma adjustment for finding noise for image": "Ajuste Sigma para encontrar ruído para imagem",
-    "Loops": "Loops",
-    "Denoising strength change factor": "Fator de mudança na quantidade do Denoise",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Configurações recomendadas: Passos de amostragem: 80-100: Euler a, força do Denoise: 0.8",
-    "Pixels to expand": "Pixels para expandir",
-    "Outpainting direction": "Direção do outpainting",
-    "left": "esquerda",
-    "right": "direita",
-    "up": "cima",
-    "down": "baixo",
-    "Fall-off exponent (lower=higher detail)": "Expoente de queda (menor=mais detalhes)",
-    "Color variation": "Variação de cor",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Amplia a imagem em dobro; ajusta a largura e altura para definir o tamanho do ladrilho",
-    "Tile overlap": "Sobreposição de ladrilho",
-    "Upscaler": "Ampliador",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR 4x": "SwinIR 4x",
-    "Single Image": "Uma imagem",
-    "Batch Process": "Processo em lote",
-    "Batch from Directory": "Lote apartir de diretório",
-    "Source": "Origem",
-    "Show result images": "Mostrar imagens resultantes",
-    "Scale by": "Aumentar proporcionalmente em",
-    "Scale to": "Aumentar proporcionalmente para",
-    "Resize": "Redimensionar",
-    "Crop to fit": "Cortar para caber",
-    "Upscaler 2 visibility": "Visibilidade da ferramenta de ampliação 2",
-    "GFPGAN visibility": "Visibilidade GFPGAN",
-    "CodeFormer visibility": "Visibilidade CodeFormer",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "Peso do CodeFormer (0 = efeito máximo, 1 = efeito mínimo)",
-    "Upscale Before Restoring Faces": "Ampliar Antes de Refinar Rostos",
-    "Send to txt2img": "Mandar para txt2img",
-    "A merger of the two checkpoints will be generated in your": "Uma fusão dos dois checkpoints será gerada em seu",
-    "checkpoint": "checkpoint",
-    "directory.": "diretório.",
-    "Primary model (A)": "Modelo primário (A)",
-    "Secondary model (B)": "Modelo secundário (B)",
-    "Tertiary model (C)": "Modelo terciário (C)",
-    "Custom Name (Optional)": "Nome personalizado (Opcional)",
-    "Multiplier (M) - set to 0 to get model A": "Multiplicador (M) - definir em 0 para obter o modelo A",
-    "Interpolation Method": "Método de Interpolação",
-    "Weighted sum": "Soma de pesos",
-    "Add difference": "Acrescentar diferença",
-    "Save as float16": "Salvar como float16",
-    "See": "Ver",
-    "wiki": "wiki",
-    "for detailed explanation.": "para explicação detalhada.",
-    "Create embedding": "Criar incorporação",
-    "Create hypernetwork": "Criar hypernetwork",
-    "Preprocess images": "Pré-processar imagens",
-    "Name": "Nome",
-    "Initialization text": "Texto de inicialização",
-    "Number of vectors per token": "Número de vetores por token",
-    "Overwrite Old Embedding": "Substituir Incorporação anterior",
-    "Modules": "Módulos",
-    "Enter hypernetwork layer structure": "Entrar na estrutura de camadas da hypernetwork",
-    "Select activation function of hypernetwork": "Selecionar a função de ativação de hypernetwork",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "tanh": "tanh",
-    "sigmoid": "sigmoid",
-    "celu": "celu",
-    "gelu": "gelu",
-    "glu": "glu",
-    "hardshrink": "hardshrink",
-    "hardsigmoid": "hardsigmoid",
-    "hardtanh": "hardtanh",
-    "logsigmoid": "logsigmoid",
-    "logsoftmax": "logsoftmax",
-    "mish": "mish",
-    "prelu": "prelu",
-    "rrelu": "rrelu",
-    "relu6": "relu6",
-    "selu": "selu",
-    "silu": "silu",
-    "softmax": "softmax",
-    "softmax2d": "softmax2d",
-    "softmin": "softmin",
-    "softplus": "softplus",
-    "softshrink": "softshrink",
-    "softsign": "softsign",
-    "tanhshrink": "tanhshrink",
-    "threshold": "threshold",
-    "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "Selecionar a inicialização de pesos de camada. relu-like - Kaiming, sigmoid-like - Xavier é recomendado",
-    "Normal": "Normal",
-    "KaimingUniform": "KaimingUniform",
-    "KaimingNormal": "KaimingNormal",
-    "XavierUniform": "XavierUniform",
-    "XavierNormal": "XavierNormal",
-    "Add layer normalization": "Adicionar normalização de camada",
-    "Use dropout": "Usar dropout",
-    "Overwrite Old Hypernetwork": "Sobrescrever Hypernetwork Anterior",
-    "Source directory": "Diretório de origem",
-    "Destination directory": "Diretório de destino",
-    "Existing Caption txt Action": "Ação de Título txt Já Existente",
-    "ignore": "ignorar",
-    "copy": "copiar",
-    "prepend": "adicionar ao início",
-    "append": "adicionar ao final",
-    "Create flipped copies": "Criar cópias espelhadas",
-    "Split oversized images into two": "Dividir imagens maiores em duas",
-    "Auto focal point crop": "Ajuste de corte em ponto focal automático",
-    "Use BLIP for caption": "Usar BLIP para o título",
-    "Use deepbooru for caption": "Usar deepbooru para o título",
-    "Split image threshold": "Limite de divisão de imagem",
-    "Split image overlap ratio": "Proporção de sobreposição da divisão de imagem",
-    "Focal point face weight": "Peso de ponto focal para rosto",
-    "Focal point entropy weight": "Peso de ponto focal para entropia",
-    "Focal point edges weight": "Peso de ponto focal para bordas",
-    "Create debug image": "Criar imagem de depuração",
-    "Preprocess": "Pré-processar",
-    "Train an embedding; must specify a directory with a set of 1:1 ratio images": "Treinar uma incorporação; precisa especificar um diretório com imagens de proporção 1:1",
-    "[wiki]": "[wiki]",
-    "Embedding": "Incorporação",
-    "Embedding Learning rate": "Taxa de aprendizagem da incorporação",
-    "Hypernetwork Learning rate": "Taxa de aprendizagem de Hypernetwork",
-    "Dataset directory": "Diretório de Dataset",
-    "Log directory": "Diretório de Log",
-    "Prompt template file": "Arquivo padrão de Prompt",
-    "Max steps": "Passos máximos",
-    "Save an image to log directory every N steps, 0 to disable": "Salvar uma imagem no diretório de log a cada N passos. 0 para desativar",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Salva uma cópia da incorporação no diretório de log a cada N passos. 0 para desativar",
-    "Save images with embedding in PNG chunks": "Salva imagens com incorporação em segmentos de PNG",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Ler parâmetros (prompt, etc...) para a aba txt2img durante os previews",
-    "Train Hypernetwork": "Treinar Hypernetwork",
-    "Train Embedding": "Treinar Incorporação",
-    "Apply settings": "Aplicar configurações",
-    "Saving images/grids": "Salvar imagens/grades",
-    "Always save all generated images": "Sempre salvar todas as imagens geradas",
-    "File format for images": "Tipo de formato das imagens salvas",
-    "Images filename pattern": "Padrão de nomeação para imagens salvas",
-    "Add number to filename when saving": "Adicionar número para o nome do arquivo quando salvar",
-    "Always save all generated image grids": "Sempre salvar todas as grades de imagens",
-    "File format for grids": "Tipo de formato das grades de imagens salvas",
-    "Add extended info (seed, prompt) to filename when saving grid": "Adicionar informações extras (seed, prompt) para os arquivos quando gerar uma grade",
-    "Do not save grids consisting of one picture": "Não salvar grades de apenas uma imagem",
-    "Prevent empty spots in grid (when set to autodetect)": "Previnir espaços vazios na grade (quando marcado para autodetectar)",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Contagem de linhas da grade; -1 para autodetectar e 0 para ser igual ao valor do tamanho das levas",
-    "Save text information about generation parameters as chunks to png files": "Salvar informações de parâmetros de geração como segmentos png",
-    "Create a text file next to every image with generation parameters.": "Criar um arquivo de texto com informações de geração junto a cada imagem gerada.",
-    "Save a copy of image before doing face restoration.": "Salva uma cópia de cada imagem antes do refinamento facial.",
-    "Quality for saved jpeg images": "Qualidade das imagens jpeg",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "Se a imagem PNG for maior que 4MB ou qualquer dimensão maior que 4000, diminuir e salvar uma cópia em JPG",
-    "Use original name for output filename during batch process in extras tab": "Usar o nome original para os arquivos de output durante o processo de levas da aba Extras",
-    "When using 'Save' button, only save a single selected image": "Quando usar o botão `Salvar`, somente salvar as imagens selecionadas.",
-    "Do not add watermark to images": "Não adicionar marca dágua nas imagens",
-    "Paths for saving": "Caminhos para salvar",
-    "Output directory for images; if empty, defaults to three directories below": "Diretório de saída para imagens; se deixado em branco, as imagens vao para os seguintes diretórios",
-    "Output directory for txt2img images": "Diretório de Saída para imagens txt2img",
-    "Output directory for img2img images": "Diretório de Saída para imagens img2img",
-    "Output directory for images from extras tab": "Diretório de Saída para a aba Extras",
-    "Output directory for grids; if empty, defaults to two directories below": "Diretório de Saída para grades; se vazio, vão para os diretórios seguintes",
-    "Output directory for txt2img grids": "Diretório de Saída para grades de imagens txt2img",
-    "Output directory for img2img grids": "Diretório de Saída para grades de imagens img2img",
-    "Directory for saving images using the Save button": "Diretório para imagens salvas utilizando o botão de salvar",
-    "Saving to a directory": "Salvando para um diretório",
-    "Save images to a subdirectory": "Salvar imagens para um subdiretório",
-    "Save grids to a subdirectory": "Salvar grades de imagens para um subdiretório",
-    "When using \"Save\" button, save images to a subdirectory": "Quando usar o botão \"Salvar\", salvar imagens para um subdiretório",
-    "Directory name pattern": "Padrão de nome de diretório",
-    "Max prompt words for [prompt_words] pattern": "Número máximo de palavras do padrão de prompt [prompt_words]",
-    "Upscaling": "Ampliando",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "Tamanho do ladrilho para ampliação ESRGAN. 0 = sem ladrilho.",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "Sobreposição de azulejo, em pixels, para amplicação  ESRGAN. Valores baixos = linhas de fusão mais aparente.",
-    "Tile size for all SwinIR.": "Tamanho do ladrilho para todo SwinIR.",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "Sobreposição de azulejo, em pixels,  para SwinIR. Valores baixos = junção mais aparente.",
-    "LDSR processing steps. Lower = faster": "Steps de processamento LDSR. Menos = rápido",
-    "Upscaler for img2img": "Ampliação para img2img",
-    "Upscale latent space image when doing hires. fix": "Ampliar a imagem do espaço latente quando usando o ajuste de alta definição - hires. fix",
-    "Face restoration": "Refinamento de rosto",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "Parâmento de peso do CodeFormer; 0 = efeito máximo; 1 = efeito mínimo",
-    "Move face restoration model from VRAM into RAM after processing": "Mover o processo de refinamento de rosto da VRAM da placa de vídeo para a RAM do computador depois do processamento.",
-    "System": "Sistema",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "Levantamento de uso de VRAM por segundo durante gerações. Deixar em 0 para desativar.",
-    "Always print all generation info to standard output": "Sempre mostrar as informações de todas as gerações no padrão de output",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Adicionar uma segunda barra de processamento no console que mostra a progressão de todo o trabalho.",
-    "Training": "Treinamento",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "Mover VAE e CLIP para a RAM quando treinando hypernetwork. Preserva VRAM.",
-    "Filename word regex": "Palavra de nome de arquivo regex",
-    "Filename join string": "Nome de arquivo join string",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Número de repetições para entrada única de imagens por época; serve apenas para mostrar o número de época",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Salvar um csv com as perdas para o diretório de log a cada N steps, 0 para desativar",
-    "Use cross attention optimizations while training": "Usar otimizações de atenção cruzada enquanto treinando",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "Checkpoints para manter no cache da RAM",
-    "Hypernetwork strength": "Força da Hypernetwork",
-     "Inpainting conditioning mask strength": "Força do inpaint para máscaras condicioniais",
-    "Apply color correction to img2img results to match original colors.": "Aplicar correção de cor nas imagens geradas em img2img, usando a imagem original como base.",
-    "Save a copy of image before applying color correction to img2img results": "Salvar uma cópia das imagens geradas em img2img antes de aplicar a correção de cor",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "Durante gerações img2img, fazer examente o número de steps definidos na barra (normalmente você faz menos steps com denoising menor).",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Ativar quantização em K samples para resultados mais nítidos e visíveis. Pode alterar seeds ja existentes. Precisa reiniciar para funcionar.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Ênfase: usar parênteses ao redor de palavras (texto de exemplo) para fazer o modelo dar mais atenção para aquela palavra ou frase, e chaves [texto de exemplo] para tirar atenção",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Usar método anterior de implementação de ênfase. Útil para reproduzir seeds antigas.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "Faz as amostragens K-diffusion produzirem imagens iguais em lotes quando criando uma única imagem",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "Aumenta a coerência por preenchimento apartir da ultima vírgula dentro de n tokens quando usando mais de 75 tokens",
-    "Filter NSFW content": "Filtra conteúdos inadequados(geralmente +18)",
-    "Stop At last layers of CLIP model": "Para na última camada do modelo CLIP",
-    "Interrogate Options": "Opções de Interrogatório",
-    "Interrogate: keep models in VRAM": "Interrogar: manter modelos na VRAM",
-    "Interrogate: use artists from artists.csv": "Interrogar: usa artistas e estilos do documento artists.csv",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interrogar: incluir classificação de tags de modelo combinando nos resultados (Não tem efeito na interrogação feita por legenda).",
-    "Interrogate: num_beams for BLIP": "Interrogar: num_beams para BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Interrogar: tamanho mínimo da descrição (tirando artistas, etc..)",
-    "Interrogate: maximum description length": "Interrogar: tamanho máximo da descrição",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: número máximo de linhas no arquivo de texto(0 = Sem limites)",
-    "Interrogate: deepbooru score threshold": "Interrogatório: limite de score deepbooru",
-    "Interrogate: deepbooru sort alphabetically": "Interrogar: organizar deepbooru por ordem alfabética",
-    "use spaces for tags in deepbooru": "usar espaços para tags em deepbooru",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "espaço (\\) colchetes em deepbooru (são usados como colchetes ao invés de dar ênfase)",
-    "User interface": "Interface de usuário",
-    "Show progressbar": "Mostrar barra de progresso",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Mostrar a criação de imagens a cada N sampling steps. Em 1 já dá para ver o processo de geração. Marcar como 0 para desativar.",
-    "Show previews of all images generated in a batch as a grid": "Mostrar previsualização de todas as imagens geradas em leva numa grade",
-    "Show grid in results for web": "Mostrar grade em resultados para web",
-    "Do not show any images in results for web": "Não mostrar nenhuma imagem em resultados para web",
-    "Add model hash to generation information": "Adicionar hash do modelo para informação de geração",
-    "Add model name to generation information": "Adicionar nome do modelo para informação de geração",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "Quando ler parâmetros de texto para a interface (de informações de PNG ou texto copiado), não alterar o modelo/intervalo selecionado.",
-    "Send seed when sending prompt or image to other interface": "Enviar seed quando enviar prompt ou imagem para outra interface",
-    "Font for image grids that have text": "Fonte para grade de imagens que têm texto",
-    "Enable full page image viewer": "Ativar visualizador de página inteira",
-    "Show images zoomed in by default in full page image viewer": "Mostrar imagens com zoom por definição no visualizador de página inteira",
-    "Show generation progress in window title.": "Mostrar barra de progresso no nome da janela.",
-    "Quicksettings list": "Lista de configurações rapidas",
-    "Localization (requires restart)": "Localização (precisa reiniciar)",
-    "ar_AR": "ar_AR",
-    "de_DE": "de_DE",
-    "es_ES": "es_ES",
-    "fr_FR": "fr_FR",
-    "it_IT": "it_IT",
-    "ja_JP": "ja_JP",
-    "ko_KR": "ko_KR",
-    "pt_BR": "pt_BR",
-    "ru_RU": "ru_RU",
-    "tr_TR": "tr_TR",
-    "zh_CN": "zh_CN",
-    "zh_TW": "zh_TW",
-    "Sampler parameters": "Parâmetros de Amostragem",
-    "Hide samplers in user interface (requires restart)": "Esconder amostragens na interface de usuário (precisa reiniciar)",
-    "eta (noise multiplier) for DDIM": "tempo estimado (multiplicador de ruído) para DDIM",
-    "eta (noise multiplier) for ancestral samplers": "tempo estimado (multiplicador de ruído) para amostragens ancestrais",
-    "img2img DDIM discretize": "Discretização de img2img DDIM",
-    "uniform": "uniforme",
-    "quad": "quad",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma noise",
-    "Eta noise seed delta": "tempo estimado para ruído seed delta",
-    "Request browser notifications": "Solicitar notificações do navegador",
-    "Download localization template": "Baixar arquivo modelo de localização",
-    "Reload custom script bodies (No ui updates, No restart)": "Recarregar scripts personalizados (Sem atualizar a interface, Sem reiniciar)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Reiniciar Gradio e atualizar componentes (Scripts personalizados, ui.py, js e css)",
-    "Installed": "Instalado",
-    "Available": "Disponível",
-    "Install from URL": "Instalado de URL",
-    "Apply and restart UI": "Apicar e reiniciar a interface",
-    "Check for updates": "Procurar por atualizações",
-    "Extension": "Extensão",
-    "URL": "URL",
-    "Update": "Atualização",
-    "Load from:": "Carregar de:",
-    "Extension index URL": "Índice de extensão URL",
-    "URL for extension's git repository": "URL para repositório git da extensão",
-    "Local directory name": "Nome do diretório local",
-    "Install": "Instalar",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "Prompt (apertar Ctrl+Enter ou Alt+Enter para gerar)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Prompt Negativo (apertar Ctrl+Enter ou Alt+Enter para gerar)",
-    "Add a random artist to the prompt.": "Adicionar um artista aleatório para o prompt.",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Lê os parâmetros de geração do prompt ou da última geraçao, caso o prompt esteja vazio.",
-    "Save style": "Salva um estilo de prompt.",
-    "Apply selected styles to current prompt": "Aplica o estilo para o prompt atual.",
-    "Stop processing current image and continue processing.": "Pula a imagem sendo gerada e vai para a próxima.",
-    "Stop processing images and return any results accumulated so far.": "Interrompe o processo e mostra o que foi gerado até então.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Estilo para aplicar; também serve para o prompt negativo e vai preencher se usado.",
-    "Do not do anything special": "Não faça nada de especial",
-    "Which algorithm to use to produce the image": "O tipo de algoritmo para gerar imagens.",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - cria mais variações para as imagens em diferentes passos. Mais que 40 passos cancela o efeito.",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit Models - Funciona melhor para inpainting.",
-    "Produce an image that can be tiled.": "Produz uma imagem que pode ser ladrilhada.",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Cria um processo em duas etapas, com uma imagem em baixa qualidade primeiro, aumenta a imagem e refina os detalhes sem alterar a composição da imagem",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Quanto o algoritmo deve manter da imagem original. Em 0, nada muda. Em 1 o algoritmo ignora a imagem original. Valores menores que 1.0 demoram mais.",
-    "How many batches of images to create": "Quantos lotes de imagens criar",
-    "How many image to create in a single batch": "Quantas imagens criar em um único lote",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - Quanto maior o valor, mais segue o prompt e quanto menor, menor segue.",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Codigo de geração de uma imagem - criando uma imagem com os mesmos parâmetros e seed trazem o mesmo resultado.",
-    "Set seed to -1, which will cause a new random number to be used every time": "Define seed como -1, deixando o valor que vai aparecer como aleatório.",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Reutilizar a seed da última geração, útil principalmente se ela foi aleatória",
-    "Seed of a different picture to be mixed into the generation.": "Seed de uma imagem diferente é misturada na geração.",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Qual a variação a ser gerada. Em 0, não tem efeito. Em 1, gera uma imagem completa com a variação de seed, (exceto com amostragens a).",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Tenta gerar uma imagem similar ao que teria sido feito com a mesma seed em dimensões especifica.",
-    "Separate values for X axis using commas.": "Separa os valores para o eixo X usando vírgulas.",
-    "Separate values for Y axis using commas.": "Separa os valores para o eixo Y usando vírgulas.",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Salva a imagem no diretório padrão ou escolhido e cria um arquivo csv com os parâmetros da geração.",
-    "Open images output directory": "Abre o diretório de saída de imagens.",
-    "How much to blur the mask before processing, in pixels.": "Transição do contorno da máscara, em pixels.",
-    "What to put inside the masked area before processing it with Stable Diffusion.":  "O que vai dentro da máscara antes de processá-la com Stable Diffusion.",
-	"fill it with colors of the image": "Preenche usando as cores da imagem.",
-    "keep whatever was there originally": "manter usando o que estava lá originalmente",
-    "fill it with latent space noise": "Preenche com ruídos do espaço latente.",
-    "fill it with latent space zeroes": "Preenche com zeros do espaço latente.",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "Faz ampliação na região com máscara para atingir a resolução desejada, faz inpainting, faz downscale para voltar à resolução original e cola na imagem original",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Redimensiona a imagem para a resolução desejada. A menos que a altura e a largura sejam iguais, você obterá uma proporção incorreta.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Redimensiona a imagem para que toda a resolução desejada seja preenchida com a imagem. Corta as partes que ficaram pra fora.",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Redimensiona a imagem para que toda a imagem esteja dentro da resolução desejada. Preenche o espaço vazio com as cores da imagem.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Número de vezes que vai repetir o processamento da imagem e usar como entrada para a próxima iteração",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "No modo de loopback, em cada loop a força do denoise é multiplicado por este valor. <1 significa diminuir a variedade para que sua sequência converta em uma imagem fixa. >1 significa aumentar a variedade para que sua sequência se torne cada vez mais caótica.",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "Para ampliação SD, quantidade de sobreposição em pixels que deve haver entre os ladrilhos. Os ladrilhos se sobrepõem para que, quando forem mesclados de volta em uma imagem, não haja linhas de junção claramente visíveis.",
-    "A directory on the same machine where the server is running.": "Um diretório na mesma máquina onde o servidor está rodando.",
-    "Leave blank to save images to the default path.": "Deixar em branco para salvar imagens no caminho padrão.",
-    "Result = A * (1 - M) + B * M": "Resultado = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Resultado = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "Primeiro e último dígito precisam ser 1. ex:`1, 2, 1`",
-    "Path to directory with input images": "Caminho para o diretório com imagens de entrada",
-    "Path to directory where to write outputs": "Caminho para o diretório para gravar as saídas",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Usa essas tags para definir como os nomes dos arquivos sao escolhidos: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; deixe em branco para manter o padrão.",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Se esta opção estiver marcada, as imagens não vão ter marca d`água. Aviso: se você não quer a marca d`água, você pode estar se envolvendo em comportamentos antiéticos",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Usa essas tags para definir como os nomes dos subdiretorios e grades são escolhidos: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; deixe em branco para manter o padrão.",
-    "Restore low quality faces using GFPGAN neural network": "Restaurar rostos de baixa qualidade usando a rede neural GFPGAN",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Esta expressão regular vai retirar palavras do nome do arquivo e serão juntadas via regex usando a opção abaixo em etiquetas usadas em treinamento. Não mexer para manter os nomes como estão.",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Esta string será usada para unir palavras divididas em uma única linha se a opção acima estiver habilitada.",
-    "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "Aplicável somente para modelos de inpaint. Determina quanto deve mascarar da imagem original para inpaint e img2img. 1.0 significa totalmente mascarado, que é o comportamento padrão. 0.0 significa uma condição totalmente não mascarada. Valores baixos ajudam a preservar a composição geral da imagem, mas vai encontrar dificuldades com grandes mudanças.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Lista de nomes de configurações, separados por vírgulas, para configurações que devem ir para a barra de acesso rápido na parte superior, em vez da guia de configuração usual. Veja modules/shared.py para nomes de configuração. Necessita reinicialização para aplicar.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Se este valor for diferente de zero, ele será adicionado à seed e usado para inicializar o RNG para ruídos ao usar amostragens com Tempo Estimado. Você pode usar isso para produzir ainda mais variações de imagens ou pode usar isso para combinar imagens de outro software se souber o que está fazendo.",
-    "Leave empty for auto": "Deixar desmarcado para automático"
-}

localizations/ru_RU.json

@@ -1,475 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "Загрузка...",
-    "view": "просмотр",
-    "api": "api",
-    "•": "•",
-    "built with gradio": "На основе Gradio",
-    "Stable Diffusion checkpoint": "Веса Stable Diffusion",
-    "txt2img": "текст-в-рисунок",
-    "img2img": "рисунок-в-рисунок",
-    "Extras": "Дополнения",
-    "PNG Info": "Информация о PNG",
-    "Image Browser": "Просмотр изображений",
-    "History": "Журнал",
-    "Checkpoint Merger": "Слияние весов",
-    "Train": "Обучение",
-    "Create aesthetic embedding": "Создать эмбеддинг эстетики",
-    "Settings": "Настройки",
-    "Prompt": "Запрос",
-    "Negative prompt": "Исключающий запрос",
-    "Run": "Запустить",
-    "Skip": "Пропустить",
-    "Interrupt": "Прервать",
-    "Generate": "Создать",
-    "Style 1": "Стиль 1",
-    "Style 2": "Стиль 2",
-    "Label": "Метка",
-    "File": "Файл",
-    "Drop File Here": "Перетащите файл сюда",
-    "-": "-",
-    "or": "или",
-    "Click to Upload": "Нажмите, чтобы загрузить",
-    "Image": "Рисунок",
-    "Check progress": "Узнать состояние",
-    "Check progress (first)": "Узнать состояние первого",
-    "Sampling Steps": "Шагов семплера",
-    "Sampling method": "Метод семплирования",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Ширина",
-    "Height": "Высота",
-    "Restore faces": "Восстановить лица",
-    "Tiling": "Замощение",
-    "Highres. fix": "HD-режим",
-    "Firstpass width": "Ширина первого прохода",
-    "Firstpass height": "Высота первого прохода",
-    "Denoising strength": "Сила шумоподавления",
-    "Batch count": "Рисунков подряд",
-    "Batch size": "Рисунков параллельно",
-    "CFG Scale": "Близость к запросу",
-    "Seed": "Семя",
-    "Extra": "Дополнения",
-    "Variation seed": "Вариация семени",
-    "Variation strength": "Вариация шумоподавления",
-    "Resize seed from width": "Поправка в семя от ширины",
-    "Resize seed from height": "Поправка в семя от высоты",
-    "Open for Clip Aesthetic!": "Clip-эстетика!",
-    "▼": "▼",
-    "Aesthetic weight": "Вес эстетики",
-    "Aesthetic steps": "Шагов эстетики",
-    "Aesthetic learning rate": "Скорость обучения эстетики",
-    "Slerp interpolation": "Slerp-интерполяция",
-    "Aesthetic imgs embedding": "Рисунки - эмбеддинги эстетики",
-    "None": "Ничего",
-    "Aesthetic text for imgs": "Имя эстетики рисунков",
-    "Slerp angle": "Угол slerp",
-    "Is negative text": "Это текст для исключения",
-    "Script": "Скрипт",
-    "Prompt matrix": "Матрица запросов",
-    "Prompts from file or textbox": "Запросы из файла или текста",
-    "X/Y plot": "X/Y-график",
-    "Put variable parts at start of prompt": "Переменное начало запроса",
-    "Show Textbox": "Показать текстовый ввод",
-    "File with inputs": "Файл входа",
-    "Prompts": "Запросы",
-    "X type": "Ось X",
-    "Nothing": "Ничего",
-    "Var. seed": "Вариация семени",
-    "Var. strength": "Вариация силы",
-    "Steps": "Число шагов",
-    "Prompt S/R": "Вариация запроса",
-    "Prompt order": "Порядок запросов",
-    "Sampler": "Семплер",
-    "Checkpoint name": "Имя файла весов",
-    "Hypernetwork": "Гиперсеть",
-    "Hypernet str.": "Строка гиперсети",
-    "Sigma Churn": "Возмущение сигмы",
-    "Sigma min": "Мин. сигма",
-    "Sigma max": "Макс. сигма",
-    "Sigma noise": "Сигма-шум",
-    "Eta": "Расчётное время",
-    "Clip skip": "Пропустить Clip",
-    "Denoising": "Шумоподавление",
-    "X values": "Значения X",
-    "Y type": "Тип Y",
-    "Y values": "Значения Y",
-    "Draw legend": "Легенда графика",
-    "Include Separate Images": "Включить отдельные рисунки",
-    "Keep -1 for seeds": "-1 для семени",
-    "Drop Image Here": "Перетащите рисунок сюда",
-    "Save": "Сохранить",
-    "Send to img2img": "В рисунок-в-рисунок",
-    "Send to inpaint": "В режим врисовывания",
-    "Send to extras": "В дополнения",
-    "Make Zip when Save?": "Создать zip при сохранении?",
-    "Textbox": "Текст",
-    "Interrogate\nCLIP": "Распознавание\nCLIP",
-    "Interrogate\nDeepBooru": "Распознавание\nDeepBooru",
-    "Inpaint": "врисовать",
-    "Batch img2img": "рисунок-в-рисунок (набор)",
-    "Image for img2img": "рисунок-в-рисунок (вход)",
-    "Image for inpainting with mask": "врисовать (вход с трафаретом)",
-    "Mask": "Трафарет",
-    "Mask blur": "Размытие трафарета",
-    "Mask mode": "Режим трафарета",
-    "Draw mask": "Нарисовать трафарет",
-    "Upload mask": "Загрузить трафарет",
-    "Masking mode": "Режим трафарета",
-    "Inpaint masked": "Внутри трафарета",
-    "Inpaint not masked": "Вне трафарета",
-    "Masked content": "Под трафаретом",
-    "fill": "залить",
-    "original": "сохранить",
-    "latent noise": "латентный шум",
-    "latent nothing": "латентная пустота",
-    "Inpaint at full resolution": "Врисовать при полном разрешении",
-    "Inpaint at full resolution padding, pixels": "Врисовать с достройкой до полного разрешения, в пикселях",
-    "Process images in a directory on the same machine where the server is running.": "Обрабатывать рисунки на том же компьютере, где сервер",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Использовать пустую папку вместо того, чтобы выводить в output",
-    "Disabled when launched with --hide-ui-dir-config.": "Выключено при запуске с  --hide-ui-dir-config",
-    "Input directory": "Папка входа",
-    "Output directory": "Папка выхода",
-    "Resize mode": "Масштабирование",
-    "Just resize": "Только сжать",
-    "Crop and resize": "Сжать и обрезать",
-    "Resize and fill": "Сжать и залить",
-    "img2img alternative test": "рисунок-в-рисунок (альтернатива)",
-    "Loopback": "Прокручивание",
-    "Outpainting mk2": "Обрисовыватель mk2",
-    "Poor man's outpainting": "Хоть какой-то обрисовыватель",
-    "SD upscale": "SD-апскейл",
-    "should be 2 or lower.": "должно быть меньше равно 2",
-    "Override `Sampling method` to Euler?(this method is built for it)": "Сменить метод семплирования на метод Эйлера?(скрипт строился с его учётом)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "Сменить `запрос` на `изначальный запрос`?(и `запрос-исключение`)",
-    "Original prompt": "Изначальный запрос",
-    "Original negative prompt": "Изначальный запрос-исключение",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "Сменить число шагов на число шагов декодирования?",
-    "Decode steps": "Шагов декодирования",
-    "Override `Denoising strength` to 1?": "Сменить силу шумоподавления на 1?",
-    "Decode CFG scale": "Близость к запросу декодирования",
-    "Randomness": "Случайность",
-    "Sigma adjustment for finding noise for image": "Поправка к сигме подбора шума для рисунка",
-    "Loops": "Циклов",
-    "Denoising strength change factor": "Множитель силы шумоподавления",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Рекоммендуемые настройки： Число шагов：80-100，Метод：Euler a，Шумоподавление：0.8",
-    "Pixels to expand": "Пикселов расширить",
-    "Outpainting direction": "Направление обрисовывания",
-    "left": "влево",
-    "right": "вправо",
-    "up": "вверх",
-    "down": "вниз",
-    "Fall-off exponent (lower=higher detail)": "Степень затухания （меньше=больше деталей）",
-    "Color variation": "Вариация цвета",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Расширит рисунок дважды； ползунки ширины и высоты устанавливают размеры плиток",
-    "Tile overlap": "Перекрытие плиток",
-    "Upscaler": "Апскейлер",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "BSRGAN 4x": "BSRGAN 4x",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR_4x": "SwinIR 4x",
-    "Single Image": "Один рисунок",
-    "Batch Process": "Набор рисунков",
-    "Batch from Directory": "Рисунки из папки",
-    "Source": "Вход",
-    "Show result images": "Показать результаты",
-    "Scale by": "Увеличить в",
-    "Scale to": "Увеличить до",
-    "Resize": "Масштабировать",
-    "Crop to fit": "Обрезать до рамки",
-    "Upscaler 2": "Апскейлер 2",
-    "Upscaler 2 visibility": "Видимость Апскейлера 2",
-    "GFPGAN visibility": "Видимость GFPGAN",
-    "CodeFormer visibility": "Видимость CodeFormer",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "Вес CodeFormer (0 = максимальное действие, 1 = минимальное)",
-    "Open output directory": "Открыть папку выхода",
-    "Send to txt2img": "В текст-в-рисунок",
-    "txt2img history": "журнал текста-в-рисунок",
-    "img2img history": "журнал рисунка-в-рисунок",
-    "extras history": "журнал дополнений",
-    "Renew Page": "Обновить страницу",
-    "extras": "дополнения",
-    "favorites": "избранное",
-    "Load": "Загрузить",
-    "Images directory": "Папка с рисунками",
-    "Prev batch": "Пред. набор",
-    "Next batch": "След. набор",
-    "First Page": "Первая страница",
-    "Prev Page": "Пред. страница",
-    "Page Index": "Список страниц",
-    "Next Page": "След. страница",
-    "End Page": "Конец страницы",
-    "number of images to delete consecutively next": "сколько рисунков удалить подряд",
-    "Delete": "Удалить",
-    "Generate Info": "Сведения о генерации",
-    "File Name": "Имя файла",
-    "Collect": "Накопить",
-    "Refresh page": "Обновить страницу",
-    "Date to": "Дата",
-    "Number": "Число",
-    "set_index": "индекс",
-    "Checkbox": "Галочка",
-    "A merger of the two checkpoints will be generated in your": "Слияние весов будет создано, где хранятся",
-    "checkpoint": "ckpt",
-    "directory.": "веса",
-    "Primary model (A)": "Первичная модель (A)",
-    "Secondary model (B)": "Вторичная модель (B)",
-    "Tertiary model (C)": "Третичная модель (C)",
-    "Custom Name (Optional)": "Произвольное имя (необязательно)",
-    "Multiplier (M) - set to 0 to get model A": "Множитель (M) - 0 даст модель A",
-    "Interpolation Method": "Метод интерполяции",
-    "Weighted sum": "Взвешенная сумма",
-    "Add difference": "Сумма разностей",
-    "Save as float16": "Сохранить как float16",
-    "See": "См.",
-    "wiki": "вики",
-    "for detailed explanation.": "для подробных объяснений.",
-    "Create embedding": "Создать эмбеддинг",
-    "Create aesthetic images embedding": "Создать эмбеддинг эстетики по рисункам",
-    "Create hypernetwork": "Создать гиперсеть",
-    "Preprocess images": "Предобработать рисунки",
-    "Name": "Имя",
-    "Initialization text": "Соответствующий текст",
-    "Number of vectors per token": "Векторов на токен",
-    "Overwrite Old Embedding": "Перезаписать эмбеддинг",
-    "Source directory": "Исходная папка",
-    "Modules": "Модули",
-    "Enter hypernetwork layer structure": "Структура слоёв гиперсети",
-    "Add layer normalization": "Добавить нормализацию слоёв",
-    "Overwrite Old Hypernetwork": "Перезаписать гиперсеть",
-    "Select activation function of hypernetwork": "Функция активации гиперсети",
-    "linear": "линейная",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "Destination directory": "Папка назначения",
-    "Existing Caption txt Action": "Что делать с предыдущим текстом",
-    "ignore": "игнорировать",
-    "copy": "копировать",
-    "prepend": "в начало",
-    "append": "в конец",
-    "Create flipped copies": "Создать отражённые копии",
-    "Split oversized images into two": "Поделить слишком большие рисунки пополам",
-    "Split oversized images": "Поделить слишком большие рисунки",
-    "Use BLIP for caption": "Использовать BLIP для названий",
-    "Use deepbooru for caption": "Использовать deepbooru для тегов",
-    "Split image threshold": "Порог разделения рисунков",
-    "Split image overlap ratio": "Пропорции разделения рисунков",
-    "Preprocess": "Предобработка",
-    "Train an embedding; must specify a directory with a set of 1:1 ratio images": "Обучить эмбеддинг； укажите папку рисунков с пропорциями 1:1",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "Обучить эмбеддинг или гиперсеть； укажите папку рисунков с пропорциями 1:1",
-    "[wiki]": "[вики]",
-    "Embedding": "Эмбеддинг",
-    "Embedding Learning rate": "Скорость обучения эмбеддинга",
-    "Hypernetwork Learning rate": "Скорость обучения гиперсети",
-    "Learning rate": "Скорость обучения",
-    "Dataset directory": "Папка датасета",
-    "Log directory": "Папка журнала",
-    "Prompt template file": "Файл шаблона запроса",
-    "Max steps": "Макс. шагов",
-    "Save an image to log directory every N steps, 0 to disable": "Сохранять рисунок каждые N шагов, 0 чтобы отключить",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Сохранять эмбеддинг каждые N шагов, 0 чтобы отключить",
-    "Save images with embedding in PNG chunks": "Сохранить рисунок с эмбеддингом в виде PNG-фрагментов",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Считать параметры (запрос и т.д.) из вкладки текст-в-рисунок для предпросмотра",
-    "Train Hypernetwork": "Обучить гиперсеть",
-    "Train Embedding": "Обучить эмбеддинг",
-    "Create an aesthetic embedding out of any number of images": "Создать эмбеддинг эстетики по любому числу рисунков",
-    "Create images embedding": "Создать эмбеддинг рисунков",
-    "Apply settings": "Применить настройки",
-    "Saving images/grids": "Сохранение рисунков/таблиц",
-    "Always save all generated images": "Всегда сохранять созданные рисунки",
-    "File format for images": "Формат файла рисунков",
-    "Images filename pattern": "Формат имени файлов рисунков",
-    "Always save all generated image grids": "Всегда сохранять созданные таблицы",
-    "File format for grids": "Формат файла таблиц",
-    "Add extended info (seed, prompt) to filename when saving grid": "Вставлять доп. сведения (семя, запрос) в имя файла таблиц",
-    "Do not save grids consisting of one picture": "Не сохранять таблицы из одного рисунка",
-    "Prevent empty spots in grid (when set to autodetect)": "Не допускать пустоты в таблицах （автообнаружение）",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Число строк таблицы； -1, чтобы автоматически, 0 — размер набора",
-    "Save text information about generation parameters as chunks to png files": "Встроить сведения о генерации в файлы png",
-    "Create a text file next to every image with generation parameters.": "Создать текстовый файл для каждого рисунка с параметрами генерации",
-    "Save a copy of image before doing face restoration.": "Сохранить копию перед восстановлением лиц",
-    "Quality for saved jpeg images": "Качество jpeg-рисунков",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "Если размер PNG больше 4МБ или рисунок шире 4000 пикселей, пересжать в JPEG",
-    "Use original name for output filename during batch process in extras tab": "Использовать исходное имя выходного файла для обработки набора во вкладке дополнений",
-    "When using 'Save' button, only save a single selected image": "Сохранять только один рисунок при нажатии кнопки Сохранить",
-    "Do not add watermark to images": "Не добавлять водяной знак",
-    "Paths for saving": "Папки сохранений",
-    "Output directory for images; if empty, defaults to three directories below": "Папка выхода рисунков; если пусто, использует те, что ниже",
-    "Output directory for txt2img images": "Папка выхода текста-в-рисунок",
-    "Output directory for img2img images": "Папка выхода рисунка-в-рисунок",
-    "Output directory for images from extras tab": "Папка выхода для дополнений",
-    "Output directory for grids; if empty, defaults to two directories below": "Папка выхода таблиц; если пусто, использует папки выше",
-    "Output directory for txt2img grids": "Папка выхода текста-в-рисунок",
-    "Output directory for img2img grids": "Папка выхода рисунка-в-рисунок",
-    "Directory for saving images using the Save button": "Папка выхода для кнопки Сохранить",
-    "Saving to a directory": "Сохранить в папку",
-    "Save images to a subdirectory": "Сохранить рисунки в подпапку",
-    "Save grids to a subdirectory": "Сохранить таблицы в подпапку",
-    "When using \"Save\" button, save images to a subdirectory": "При нажатии кнопки Сохранить, сложить рисунки в подпапку",
-    "Directory name pattern": "Шаблон имени папки",
-    "Max prompt words for [prompt_words] pattern": "Макс. число слов для шаблона [prompt_words]",
-    "Upscaling": "Апскейл",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "Размер плитки для ESRGAN. 0 = нет замощения",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "Наложение плиток ESRGAN, в пикселях. Меньше = выделеннее шов",
-    "Tile size for all SwinIR.": "Размер плиток SwinIR",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "Наложение плиток SwinIR, в пикселях. Меньше = выделеннее шов",
-    "LDSR processing steps. Lower = faster": "Число шагов LDSR. Меньше = быстрее",
-    "Upscaler for img2img": "Апскейлер рисунка-в-рисунок",
-    "Upscale latent space image when doing hires. fix": "Апскейлить образ латентного пространства для HD-режима",
-    "Face restoration": "Восстановление лиц",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "Вес CodeFormer； 0 = максимальное действие； 1 = минимальное",
-    "Move face restoration model from VRAM into RAM after processing": "Переместить модель восстановления лиц из ВОЗУ в ОЗУ после обработки",
-    "System": "Система",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "Сколько раз в секунду следить за потреблением ВОЗУ. 0, чтобы отключить",
-    "Always print all generation info to standard output": "Выводить все сведения о генерации в стандартный вывод",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Вторая шкала прогресса для всей задачи",
-    "Training": "Обучение",
-    "Unload VAE and CLIP from VRAM when training": "Убрать VAE и CLIP из ВОЗУ на время обучения",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "Переместить VAE и CLIP в ОЗУ на время обучения гиперсети. Сохраняет ВОЗУ",
-    "Filename word regex": "Regex имени файла",
-    "Filename join string": "Дополнить к имени файла",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Число повторов для каждого рисунка за эпоху; используется только, чтобы отобразить число эпохи",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Сохранять csv с параметром loss в папку журнала каждые N шагов, 0 - отключить",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "Удерживать веса в ОЗУ",
-    "Hypernetwork strength": "Сила гиперсети",
-    "Apply color correction to img2img results to match original colors.": "Цветокоррекция вывода рисунка-в-рисунок, сохраняющая исходные цвета",
-    "Save a copy of image before applying color correction to img2img results": "Сохранить копию рисунка перед цветокоррекцией",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "В режиме рисунок-в-рисунок сделать ровно указанное ползунком число шагов (обычно шумоподавление их уменьшает)",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Включить квантование К-семплерах для более резких и чистых результатов. Может потребовать поменять семя. Требует перезапуска.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Скобки: (понятие) - больше внимания к тексту, [понятие] - меньше внимания к тексту",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Включить старую обработку скобок. Может потребоваться, чтобы воспроизвести старые семена.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "Заставить семплеры K-diffusion производить тот же самый рисунок в наборе, как и в единичной генерации",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "Увеличить связность, достраивая запрос от последней запятой до n токенов, когда используется свыше 75 токенов",
-    "Filter NSFW content": "Фильтровать небезопасный контент",
-    "Stop At last layers of CLIP model": "Остановиться на последних слоях модели CLIP",
-    "Interrogate Options": "Опции распознавания",
-    "Interrogate: keep models in VRAM": "Распознавание: хранить модели в ВОЗУ",
-    "Interrogate: use artists from artists.csv": "Распознавание: использовать художников из artists.csv",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Распознавание: включить ранжирование совпавших тегов в результате (не работает для распознавателей-создателей заголовков)",
-    "Interrogate: num_beams for BLIP": "Распознавание: num_beams для BLIP",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Распознавание: минимальная длина описания (исключая художников и т.п.)",
-    "Interrogate: maximum description length": "Распознавание: максимальная длина описания",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: максимальное число строк в текстовом файле (0 = без ограничений)",
-    "Interrogate: deepbooru score threshold": "Распознавание: ограничение счёта deepbooru",
-    "Interrogate: deepbooru sort alphabetically": "Распознавание: сортировать deepbooru по алфавиту",
-    "use spaces for tags in deepbooru": "Пробелы для тегов deepbooru",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "Использовать скобки в deepbooru как обычные скобки, а не для усиления",
-    "User interface": "Пользовательский интерфейс",
-    "Show progressbar": "Шкала прогресса",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Показывать процесс созданния рисунка каждые N шагов. 0 - отключить",
-    "Show grid in results for web": "Показать таблицу в выводе браузера",
-    "Do not show any images in results for web": "Не показывать выходные рисунки в браузере",
-    "Add model hash to generation information": "Добавить хеш весов к параметрам генерации",
-    "Add model name to generation information": "Добавить имя весов к параметрам генерации",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "При считывании параметров генерации из текста в интерфейс, не менять выбранную модель/веса.",
-    "Font for image grids that have text": "Шрифт для таблиц, содержащих текст",
-    "Enable full page image viewer": "Включить полноэкранный просмотр картинок",
-    "Show images zoomed in by default in full page image viewer": "По умолчанию увеличивать картинки в полноэкранном просмотре",
-    "Show generation progress in window title.": "Отображать прогресс в имени вкладки",
-    "Quicksettings list": "Список быстрых настроек",
-    "Localization (requires restart)": "Перевод （требует перезапуск）",
-    "Sampler parameters": "Параметры семплера",
-    "Hide samplers in user interface (requires restart)": "Убрать семплеры из интерфейса (требует перезапуск)",
-    "eta (noise multiplier) for DDIM": "eta (множитель шума) DDIM",
-    "eta (noise multiplier) for ancestral samplers": "eta (множитель шума) для ancestral-семплеров",
-    "img2img DDIM discretize": "дискретизация DDIM для рисунка-в-рисунок",
-    "uniform": "однородная",
-    "quad": "квадратичная",
-    "sigma churn": "сигма-вариация",
-    "sigma tmin": "сигма-tmin",
-    "sigma noise": "сигма-шум",
-    "Eta noise seed delta": "Eta (дельта шума семени)",
-    "Images Browser": "Просмотр изображений",
-    "Preload images at startup": "Предзагружать рисунки во время запуска",
-    "Number of pictures displayed on each page": "Число рисунков на каждой странице",
-    "Minimum number of pages per load": "Мин. число загружаемых страниц",
-    "Number of grids in each row": "Число таблиц в каждой строке",
-    "Request browser notifications": "Запросить уведомления браузера",
-    "Download localization template": "Загрузить щаблон перевода",
-    "Reload custom script bodies (No ui updates, No restart)": "Перезагрузить пользовательские скрипты (не требует обновления интерфейса и перезапуска)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Перезагрузить Gradio и обновить компоненты (только пользовательские скрипты, ui.py, js и css)",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "Запрос （нажмите Ctrl+Enter или Alt+Enter для генерации）",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Запрос-исключение （нажмите Ctrl+Enter или Alt+Enter для генерации）",
-    "Add a random artist to the prompt.": "Добавить случайного художника к запросу",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Считать параметры генерации из запроса или из предыдущей генерации в пользовательский интерфейс, если пусто",
-    "Save style": "Сохранить стиль",
-    "Apply selected styles to current prompt": "Применить выбранные стили к текущему промпту",
-    "Stop processing current image and continue processing.": "Прекратить обрабатывать текущий рисунок, но продолжить работу",
-    "Stop processing images and return any results accumulated so far.": "Прекратить обрабатку рисунков и вернуть всё, что успели сделать.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Стиль к применению; стили содержат как запрос, так и исключение, и применяют их оба",
-    "Do not do anything special": "Не делать ничего особенного",
-    "Which algorithm to use to produce the image": "Какой алгоритм использовать для того, чтобы произвести рисунок",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - очень творческий, в зависимости от числа шагов может привести совершенно к различным результатам, выше 30-40 лучше не ставить",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit модели - лучше всего для обрисовки",
-    "Produce an image that can be tiled.": "Сделать из рисунка непрерывную обёртку",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Применить двушаговый процесс, чтобы создать рисунок на меньшем разрешении, апскейлнуть, а затем улучшить детали без смены композиции",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Определяет, насколько сильно алгоритм будет опираться на содержание изображения. 0 - не меняет ничего, 1 - совсем не связанный выход. Меньше 1.0 процесс использует меньше шагов, чем указано их ползунком.",
-    "How many batches of images to create": "Сколько создать наборов из картинок",
-    "How many image to create in a single batch": "Сколько картинок создать в каждом наборе",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale: насколько сильно изображение должно соответсвтовать запросу — меньшие значения приведут к более свободным итогам",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Значение, которое определяет выход генератора случайных чисел — если вы создадите рисунок с теми же параметрами и семенем, как у другого изображения, вы получите тот же результат",
-    "Set seed to -1, which will cause a new random number to be used every time": "Установить семя в -1, что вызовет каждый раз случайное число",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Использовать семя предыдущей генерации, обычно полезно, если оно было случайным",
-    "Seed of a different picture to be mixed into the generation.": "Семя с другого рисунка, подмешенного в генерацию.",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Насколько сильную вариацию произвести. При 0м значении действия не будет. Для 1 вы получите полноценный рисунок с семенем вариации (кроме ancestral-семплеров, где вы просто что-то получите).",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Попытаться воспроизвести изображение, похожее на то, чтобы получилось с тем же семенем на выбранном разрешении",
-    "This text is used to rotate the feature space of the imgs embs": "Этот текст используется, чтобы произвести вращение пространства признаков из эмбеддинга рисунков",
-    "Separate values for X axis using commas.": "Отдельные значения оси X через запятую.",
-    "Separate values for Y axis using commas.": "Отдельные значения оси Y через запятую.",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Записать изображение в папку (по-умолчанию - log/images), а параметры генерации - в csv файл",
-    "Open images output directory": "Открыть папку сохранения изображений",
-    "How much to blur the mask before processing, in pixels.": "Насколько пикселей размыть трафарет перед обработкой",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Что поместить в область под трафаретом перед обработкой Stable Diffusion",
-    "fill it with colors of the image": "залить цветами изображения",
-    "keep whatever was there originally": "сохранить то, что было до этого",
-    "fill it with latent space noise": "залить латентным шумом",
-    "fill it with latent space zeroes": "залить латентными нулями",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "апскейл до нужного разрешения, врисовка, сжатие до начального размера и вставка в исходный рисунок",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Масшабировать изображение до нужного разрешения. Если только высота и ширина не совпадают, вы получите неверное соотношение сторон.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Масштабировать изображение так, чтобы им заполнялось всё выбранное выходное разрешение. Обрезать выступающие части",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Масштабировать изображение так, всё изображение помещалось в выбранное выходное разрешение. Заполнить пустое место цветами изображения.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Сколько раз повторить обработку изображения и использовать её как вход для следующией итерации",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "В режиме прокрутки, для каждого цикла сила шумоподавления умножается на это значение. <1 уменьшает вариации так, чтобы последовательность сошлась на какой-то одной картинке. >1 увеличивает вариации, так что ваша последовательность станет всё более и более сумбурной.",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "Для SD-апскейла, как много перекрытия в пикселях должно быть между плитками. Плитки перекрываются таким образом, чтобы они могли сойтись обратно в единое изображение, без видимого шва.",
-    "A directory on the same machine where the server is running.": "Папка на той же машине, где запущен сервер",
-    "Leave blank to save images to the default path.": "Оставьте пустым, чтобы сохранить рисунки в папку по-умолчанию",
-    "Result = A * (1 - M) + B * M": "Выход = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Выход = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "1я и последняя цифры должны быть 1. напр.'1, 2, 1'",
-    "how fast should the training go. Low values will take longer to train, high values may fail to converge (not generate accurate results) and/or may break the embedding (This has happened if you see Loss: nan in the training info textbox. If this happens, you need to manually restore your embedding from an older not-broken backup).\n\nYou can set a single numeric value, or multiple learning rates using the syntax:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nWill train with rate of 0.005 for first 100 steps, then 1e-3 until 1000 steps, then 1e-5 for all remaining steps.": "как быстро будет происходить обучение. Меньшие значения увеличат время обучения, но высокие могут нарушить сходимость модели (не будет создавать должные результаты) и/или сломать эмбеддинг. (Это случилось, если вы видете Loss: nan в текстовом окне вывода обучения. В этом случае вам придётся восстанавливать эмбеддинг вручную из старой, не повреждённой резервной копии).\n\nВы также можете указать единичное значение или последовательность из нескольких, используя следующий синтаксис:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nБудет обучаться со скоростью 0.005 первые 100 шагов, затем 1e-3 до 1000 шагов, после 1e-5 для всех оставшихся шагов.",
-    "Path to directory with input images": "Путь к папке со входными изображениями",
-    "Path to directory where to write outputs": "Путь к папке, в которую записывать результаты",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "Используйте следующие теги, чтобы определить, как подбираются названия файлов для изображений: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; если пусто, используется значение по-умолчанию",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Когда эта опция включена, на созданные изображения не будет добавляться водяной знак. Предупреждение: не добавляя водяной знак, вы, вероятно, ведёте себя аморально.",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; leave empty for default.": "Используйте следующие теги, чтобы определить, как подбираются названия подпапок для рисунков и табоиц: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [job_timestamp]; если пусто, используется значение по-умолчанию",
-    "Restore low quality faces using GFPGAN neural network": "Восстановить низкокачественные лица, используя нейросеть GFPGAN",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Это регулярное выражение будет использовано, чтобы извлечь слова из имени файла, и они будут соединены с текстом в метке ниже как вход во время обучения. Оставьте пустым, чтобы сохранить имя файла как есть",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Эта строка будет использована, чтобы объединить разделённые слова в одну строку, если включена опция выше.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Список имён настроек, разделённый запятыми, предназначенных для быстрого доступа через панель наверху, а не через привычную вкладку настроек. Для применения требует перезапуска.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Если это значение не нулевое, оно будет добавлено к семени и использовано для инициалицации ГСЧ шума семплеров с параметром Eta. Вы можете использовать это, чтобы произвести ещё больше вариаций рисунков, либо же для того, чтобы подойти близко к результатам других программ, если знаете, что делаете.",
-    "Enable Autocomplete": "Включить автодополнение",
-    "Allowed categories for random artists selection when using the Roll button": "Разрешённые категории художников для случайного выбора при использовании кнопки + три",
-    "Roll three": "+ три",
-    "Generate forever": "Непрерывная генерация",
-    "Cancel generate forever": "Отключить непрерывную генерацию"
-}

localizations/tr_TR.json

@@ -1,423 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "Yükleniyor...",
-    "view": "arayüz",
-    "api": "",
-    "•": "-",
-    "built with gradio": "gradio ile inşa edildi",
-    "Stable Diffusion checkpoint": "Kararlı Difüzyon kontrol noktası",
-    "txt2img": "txt2img",
-    "img2img": "img2img",
-    "Extras": "Ekstralar",
-    "PNG Info": "PNG Bilgisi",
-    "Checkpoint Merger": "Checkpoint Birleştir",
-    "Train": "Eğitim",
-    "Settings": "Ayarlar",
-    "Prompt": "İstem",
-    "Negative prompt": "Negatif istem",
-    "Run": "Koşmak",
-    "Skip": "Atla",
-    "Interrupt": "Durdur",
-    "Generate": "Oluştur",
-    "Style 1": "Stil 1",
-    "Style 2": "Stil 2",
-    "Label": "Etiket",
-    "File": "Dosya",
-    "Drop File Here": "Dosyayı Buraya Bırakın",
-    "-": "-",
-    "or": "veya",
-    "Click to Upload": "Yüklemek için Tıklayınız",
-    "Image": "Resim",
-    "Check progress": "İlerlemeyi kontrol edin",
-    "Check progress (first)": "Önce ilerlemeyi kontrol edin",
-    "Sampling Steps": "Örnekleme Adımları",
-    "Sampling method": "Örnekleme yöntemi",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM hızlı",
-    "DPM adaptive": "DPM uyarlanabilir",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "Genişlik",
-    "Height": "Yükseklik",
-    "Restore faces": "Yüzleri düzeltme",
-    "Tiling": "Döşeme Oluştur",
-    "Highres. fix": "Highres. düzeltme",
-    "Firstpass width": "İlk geçiş genişliği",
-    "Firstpass height": "İlk geçiş yüksekliği",
-    "Denoising strength": "Gürültü arındırma gücü",
-    "Batch count": "Grup sayısı",
-    "Batch size": "Grup büyüklüğü",
-    "CFG Scale": "CFG Ölçeği",
-    "Seed": "Tohum",
-    "Extra": "Ekstra",
-    "Variation seed": "Varyasyon tohumu",
-    "Variation strength": "Varyasyon gücü",
-    "Resize seed from width": "Tohumu genişlik ile yeniden boyutlandırma",
-    "Resize seed from height": "Tohumu yükseklik ile yeniden boyutlandırma",
-    "Script": "Scriptler",
-    "None": "Hiçbiri",
-    "Prompt matrix": "İstem matrisi",
-    "Prompts from file or textbox": "Dosyadan veya metin kutusundan istemler",
-    "X/Y plot": "X/Y grafiği",
-    "Put variable parts at start of prompt": "Değişken parçaları komut isteminin başına koyun",
-    "Show Textbox": "Metin Kutusunu Göster",
-    "File with inputs": "Girdileri içeren dosya",
-    "Prompts": "İpuçları",
-    "X type": "X tipi",
-    "Nothing": "Hiçbir şey",
-    "Var. seed": "Var. tohum",
-    "Var. strength": "Var. güç",
-    "Steps": "Adımlar",
-    "Prompt S/R": "İstem S/R",
-    "Prompt order": "İstem sırası",
-    "Sampler": "Örnekleyici",
-    "Checkpoint name": "Kontrol noktası adı",
-    "Hypernetwork": "Hipernetwork",
-    "Hypernet str.": "Hypernet str.",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "Sigma dakika",
-    "Sigma max": "Sigma maksimum",
-    "Sigma noise": "Sigma gürültüsü",
-    "Eta": "Eta",
-    "Clip skip": "Klip atlama",
-    "Denoising": "Denoising",
-    "X values": "X değerleri",
-    "Y type": "Y tipi",
-    "Y values": "Y değerleri",
-    "Draw legend": "Gösterge çizin",
-    "Include Separate Images": "Ayrı Görseller Ekleyin",
-    "Keep -1 for seeds": "Tohumlar için -1'i saklayın",
-    "Drop Image Here": "Resmi Buraya Bırakın",
-    "Save": "Kaydet",
-    "Send to img2img": "img2img'ye gönder",
-    "Send to inpaint": "Inpaint'e gönder",
-    "Send to extras": "Ekstralara gönder",
-    "Make Zip when Save?": "Kaydederken Zip Yap?",
-    "Textbox": "Metin Kutusu",
-    "Interrogate\nCLIP": "Sorgula\nCLIP",
-    "Inpaint": "Inpaint",
-    "Batch img2img": "Toplu img2img",
-    "Image for img2img": "img2img için resim",
-    "Image for inpainting with mask": "Maske ile inpainting için görüntü",
-    "Mask": "Maske",
-    "Mask blur": "Maske bulanıklığı",
-    "Mask mode": "Maske modu",
-    "Draw mask": "Maske çizin",
-    "Upload mask": "Maske yükle",
-    "Masking mode": "Maskeleme modu",
-    "Inpaint masked": "Maskeli inpaint",
-    "Inpaint not masked": "Boya maskelenmemiş",
-    "Masked content": "Maskelenmiş içerik",
-    "fill": "doldurun",
-    "original": "orijinal",
-    "latent noise": "gizli gürültü",
-    "latent nothing": "gizli hiçbir şey",
-    "Inpaint at full resolution": "Tam çözünürlükte inpaint",
-    "Inpaint at full resolution padding, pixels": "Tam çözünürlükte inpaint dolgu, piksel",
-    "Process images in a directory on the same machine where the server is running.": "Görüntüleri sunucunun çalıştığı makinedeki bir dizinde işleyin.",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "Resimleri çıktı dizinine yazmak yerine normal şekilde kaydetmek için boş bir çıktı dizini kullanın.",
-    "Input directory": "Girdi dizini",
-    "Output directory": "Çıktı dizini",
-    "Resize mode": "Yeniden boyutlandırma modu",
-    "Just resize": "Sadece yeniden boyutlandır",
-    "Crop and resize": "Kırpma ve yeniden boyutlandırma",
-    "Resize and fill": "Yeniden boyutlandırın ve doldurun",
-    "img2img alternative test": "img2img alternatif test",
-    "Loopback": "Geri Döngü",
-    "Outpainting mk2": "Outpainting mk2",
-    "Poor man's outpainting": "Zavallı adamın dış boyaması",
-    "SD upscale": "SD lüks",
-    "should be 2 or lower.": "2 veya daha düşük olmalıdır.",
-    "Override `Sampling method` to Euler?(this method is built for it)": "Euler için `Örnekleme yöntemini` geçersiz kılın (bu yöntem bunun için oluşturulmuştur)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "Prompt` değerini `orijinal prompt` ile aynı değere geçersiz kılma (ve `negatif prompt`)",
-    "Original prompt": "Orijinal bilgi istemi",
-    "Original negative prompt": "Orijinal negatif istem",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "Örnekleme Adımlarını `Kod çözme adımları` ile aynı değere mi geçersiz kılıyorsunuz?",
-    "Decode steps": "Kod çözme adımları",
-    "Override `Denoising strength` to 1?": "`Denoising strength` değerini 1 olarak geçersiz kıl?",
-    "Decode CFG scale": "CFG ölçeğinin kodunu çöz",
-    "Randomness": "Rastgelelik",
-    "Sigma adjustment for finding noise for image": "Görüntü için gürültü bulmaya yönelik Sigma ayarı",
-    "Loops": "Döngüler",
-    "Denoising strength change factor": "Denoising gücü değişim faktörü",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "Önerilen ayarlar: Örnekleme Adımları: 80-100, Örnekleyici: Euler a, Denoising gücü: 0.8",
-    "Pixels to expand": "Genişletilecek pikseller",
-    "Outpainting direction": "Dış boyama yönü",
-    "left": "sol",
-    "right": "doğru",
-    "up": "yukarı",
-    "down": "aşağı",
-    "Fall-off exponent (lower=higher detail)": "Düşme üssü (düşük=daha yüksek detay)",
-    "Color variation": "Renk çeşitliliği",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "Görüntüyü boyutlarının iki katına yükseltir; döşeme boyutunu ayarlamak için genişlik ve yükseklik kaydırıcılarını kullanın",
-    "Tile overlap": "Karo örtüşmesi",
-    "Upscaler": "Upscaler",
-    "Lanczos": "Lanczos",
-    "LDSR": "LDSR",
-    "SwinIR 4x": "SwinIR 4x",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "Single Image": "Tek Resim",
-    "Batch Process": "Toplu İşlem",
-    "Batch from Directory": "Dizinden Toplu İş",
-    "Source": "Kaynak",
-    "Show result images": "Sonuç resimlerini göster",
-    "Scale by": "Ölçek tarafından",
-    "Scale to": "Ölçeklendir",
-    "Resize": "Yeniden Boyutlandır",
-    "Crop to fit": "Sığdırmak için kırpın",
-    "Upscaler 2 visibility": "Upscaler 2 görünürlüğü",
-    "GFPGAN visibility": "GFPGAN görünürlüğü",
-    "CodeFormer visibility": "CodeFormer görünürlüğü",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormer ağırlığı (0 = maksimum etki, 1 = minimum etki)",
-    "Open output directory": "Çıktı dizinini aç",
-    "Send to txt2img": "txt2img'ye gönder",
-    "A merger of the two checkpoints will be generated in your": "İki kontrol noktasının bir birleşimi sizin kontrol noktanızda oluşturulacaktır.",
-    "checkpoint": "kontrol noktası",
-    "directory.": "dizin.",
-    "Primary model (A)": "Birincil model (A)",
-    "Secondary model (B)": "İkincil model (B)",
-    "Tertiary model (C)": "Üçüncü model (C)",
-    "Custom Name (Optional)": "Özel Ad (İsteğe Bağlı)",
-    "Multiplier (M) - set to 0 to get model A": "Çarpan (M) - A modelini elde etmek için 0'a ayarlayın",
-    "Interpolation Method": "İnterpolasyon Yöntemi",
-    "Weighted sum": "Ağırlıklı toplam",
-    "Add difference": "Farklılık ekleyin",
-    "Save as float16": "float16 olarak kaydet",
-    "See": "Bkz. ",
-    "wiki": "wiki",
-    "for detailed explanation.": " ayrıntılı açıklama için.",
-    "Create embedding": "Yerleştirme oluşturma",
-    "Create hypernetwork": "Hipernet oluşturun",
-    "Preprocess images": "Görüntüleri ön işleme",
-    "Name": "İsim",
-    "Initialization text": "Başlatma metni",
-    "Number of vectors per token": "Belirteç başına vektör sayısı",
-    "Overwrite Old Embedding": "Eski Yerleştirmenin Üzerine Yaz",
-    "Modules": "Modüller",
-    "Enter hypernetwork layer structure": "Hipernetwork katman yapısına girin",
-    "Select activation function of hypernetwork": "Hipernetwork'ün aktivasyon fonksiyonunu seçin",
-    "linear": "doğrusal",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "Add layer normalization": "Katman normalizasyonu ekleyin",
-    "Use dropout": "Bırakmayı kullanın",
-    "Overwrite Old Hypernetwork": "Eski Hipernetwork'ün Üzerine Yazma",
-    "Source directory": "Kaynak dizini",
-    "Destination directory": "Hedef dizini",
-    "Existing Caption txt Action": "Mevcut Başlık txt Eylem",
-    "ignore": "görmezden gel",
-    "copy": "kopya",
-    "prepend": "prepend",
-    "append": "ekle",
-    "Create flipped copies": "Ters çevrilmiş kopyalar oluşturun",
-    "Split oversized images": "Büyük boyutlu görüntüleri bölme",
-    "Use BLIP for caption": "Başlık için BLIP kullanın",
-    "Use deepbooru for caption": "Başlık için deepbooru kullanın",
-    "Split image threshold": "Bölünmüş görüntü eşiği",
-    "Split image overlap ratio": "Bölünmüş görüntü örtüşme oranı",
-    "Preprocess": "Ön işlem",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "Bir gömme veya Hipernetwork eğitin; 1:1 oranlı görüntülerin bulunduğu bir dizin belirtmelisiniz",
-    "[wiki]": "[wiki]",
-    "Embedding": "Yerleştirme",
-    "Embedding Learning rate": "Gömme Öğrenme oranı",
-    "Hypernetwork Learning rate": "Hypernetwork Öğrenme oranı",
-    "Dataset directory": "Veri seti dizini",
-    "Log directory": "Günlük dizini",
-    "Prompt template file": "Komut istemi şablon dosyası",
-    "Max steps": "Maksimum adım",
-    "Save an image to log directory every N steps, 0 to disable": "Her N adımda bir görüntüyü günlük dizinine kaydet, 0 devre dışı bırakmak için",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "Katıştırmanın bir kopyasını her N adımda bir günlük dizinine kaydedin, devre dışı bırakmak için 0",
-    "Save images with embedding in PNG chunks": "Görüntüleri PNG parçalarına yerleştirerek kaydedin",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "Önizleme yaparken txt2img sekmesinden parametreleri (istem, vb...) okuma",
-    "Train Hypernetwork": "Tren Hipernetwork",
-    "Train Embedding": "Tren Gömme",
-    "Apply settings": "Ayarları uygula",
-    "Saving images/grids": "Görüntüleri/gridleri kaydetme",
-    "Always save all generated images": "Oluşturulan tüm görüntüleri her zaman kaydedin",
-    "File format for images": "Görüntüler için dosya formatı",
-    "Images filename pattern": "Görüntü dosya adı deseni",
-    "Add number to filename when saving": "Kaydederken dosya adına numara ekle",
-    "Always save all generated image grids": "Oluşturulan tüm görüntü ızgaralarını her zaman kaydedin",
-    "File format for grids": "Izgaralar için dosya formatı",
-    "Add extended info (seed, prompt) to filename when saving grid": "Izgarayı kaydederken dosya adına genişletilmiş bilgi (tohum, istem) ekleyin",
-    "Do not save grids consisting of one picture": "Tek resimden oluşan ızgaraları kaydetmeyin",
-    "Prevent empty spots in grid (when set to autodetect)": "Izgaradaki boş noktaları önleme (otomatik algılamaya ayarlandığında)",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "Izgara satır sayısı; otomatik algılama için -1, yığın boyutuyla aynı olması için 0 kullanın",
-    "Save text information about generation parameters as chunks to png files": "Üretim parametreleri hakkındaki metin bilgilerini png dosyalarına parçalar halinde kaydedin",
-    "Create a text file next to every image with generation parameters.": "Oluşturma parametreleri ile her görüntünün yanında bir metin dosyası oluşturun.",
-    "Save a copy of image before doing face restoration.": "Yüz restorasyonu yapmadan önce görüntünün bir kopyasını kaydedin.",
-    "Quality for saved jpeg images": "Kaydedilen jpeg görüntüleri için kalite",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "PNG görüntüsü 4MB'den büyükse veya herhangi bir boyut 4000'den büyükse, ölçeği küçültün ve kopyayı JPG olarak kaydedin",
-    "Use original name for output filename during batch process in extras tab": "Ekstralar sekmesinde toplu işlem sırasında çıktı dosya adı için orijinal adı kullan",
-    "When using 'Save' button, only save a single selected image": "'Kaydet' düğmesini kullanırken, yalnızca seçilen tek bir resmi kaydedin",
-    "Do not add watermark to images": "Görüntülere filigran eklemeyin",
-    "Paths for saving": "Tasarruf için yollar",
-    "Output directory for images; if empty, defaults to three directories below": "Görüntüler için çıktı dizini; boşsa, varsayılan olarak aşağıdaki üç dizine gider",
-    "Output directory for txt2img images": "txt2img görüntüleri için çıktı dizini",
-    "Output directory for img2img images": "img2img görüntüleri için çıktı dizini",
-    "Output directory for images from extras tab": "Ekstralar sekmesindeki görüntüler için çıktı dizini",
-    "Output directory for grids; if empty, defaults to two directories below": "Izgaralar için çıktı dizini; boşsa, varsayılan olarak aşağıdaki iki dizine gider",
-    "Output directory for txt2img grids": "txt2img ızgaraları için çıktı dizini",
-    "Output directory for img2img grids": "img2img ızgaraları için çıktı dizini",
-    "Directory for saving images using the Save button": "Kaydet düğmesini kullanarak görüntüleri kaydetmek için dizin",
-    "Saving to a directory": "Bir dizine kaydetme",
-    "Save images to a subdirectory": "Görüntüleri bir alt dizine kaydetme",
-    "Save grids to a subdirectory": "Izgaraları bir alt dizine kaydetme",
-    "When using \"Save\" button, save images to a subdirectory": "\"Kaydet\" düğmesini kullanırken, görüntüleri bir alt dizine kaydedin",
-    "Directory name pattern": "Dizin adı kalıbı",
-    "Max prompt words for [prompt_words] pattern": "prompt_words] kalıbı için maksimum istem sözcükleri",
-    "Upscaling": "Yükseltme",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "ESRGAN yükselticileri için döşeme boyutu. 0 = döşeme yok.",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "ESRGAN yükselticileri için piksel cinsinden döşeme örtüşmesi. Düşük değerler = görünür bağlantı hattı.",
-    "Tile size for all SwinIR.": "Tüm SwinIR için döşeme boyutu.",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "SwinIR için piksel cinsinden döşeme örtüşmesi. Düşük değerler = görünür dikiş.",
-    "LDSR processing steps. Lower = faster": "LDSR işleme adımları. Düşük = daha hızlı",
-    "Upscaler for img2img": "img2img için üst ölçekleyici",
-    "Upscale latent space image when doing hires. fix": "İşe alım yaparken gizli alan görüntüsünü yükselt. düzelt",
-    "Face restoration": "Yüz restorasyonu",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormer ağırlık parametresi; 0 = maksimum etki; 1 = minimum etki",
-    "Move face restoration model from VRAM into RAM after processing": "İşlemden sonra yüz restorasyon modelini VRAM'den RAM'e taşıma",
-    "System": "Sistem",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "Üretim sırasında saniye başına VRAM kullanım yoklamaları. Devre dışı bırakmak için 0 olarak ayarlayın.",
-    "Always print all generation info to standard output": "Tüm üretim bilgilerini her zaman standart çıktıya yazdır",
-    "Add a second progress bar to the console that shows progress for an entire job.": "Konsola tüm iş için ilerlemeyi gösteren ikinci bir ilerleme çubuğu ekleyin.",
-    "Training": "Eğitim",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "Hiperneti eğitirken VAE ve CLIP'i RAM'e taşıyın. VRAM'den tasarruf sağlar.",
-    "Filename word regex": "Dosya adı kelime regex",
-    "Filename join string": "Dosya adı birleştirme dizesi",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "Epok başına tek bir girdi görüntüsü için tekrar sayısı; yalnızca epok numarasını görüntülemek için kullanılır",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "Her N adımda bir günlük dizinine kaybı içeren bir csv kaydedin, devre dışı bırakmak için 0",
-    "Stable Diffusion": "Kararlı Difüzyon",
-    "Checkpoints to cache in RAM": "RAM'de önbelleğe alınacak kontrol noktaları",
-    "Hypernetwork strength": "Hipernetwork gücü",
-    "Apply color correction to img2img results to match original colors.": "Orijinal renklerle eşleştirmek için img2img sonuçlarına renk düzeltmesi uygulayın.",
-    "Save a copy of image before applying color correction to img2img results": "img2img sonuçlarına renk düzeltmesi uygulamadan önce görüntünün bir kopyasını kaydedin",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "img2img ile, kaydırıcının belirttiği adım miktarını tam olarak yapın (normalde daha az denoising ile daha az yaparsınız).",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "Daha keskin ve temiz sonuçlar için K örnekleyicilerinde nicelemeyi etkinleştirin. Bu, mevcut tohumları değiştirebilir. Uygulamak için yeniden başlatma gerektirir.",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "Vurgu: modelin metne daha fazla dikkat etmesini sağlamak için (metin) ve daha az dikkat etmesini sağlamak için [metin] kullanın",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "Eski vurgu uygulamasını kullanın. Eski tohumları yeniden üretmek faydalı olabilir.",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "K-difüzyon örnekleyicilerinin tek bir görüntü oluştururken olduğu gibi toplu halde aynı görüntüleri üretmesini sağlayın",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "75'ten fazla belirteç kullanıldığında n belirteç içindeki son virgülden itibaren dolgu yaparak tutarlılığı artırın",
-    "Filter NSFW content": "NSFW içeriği filtreleme",
-    "Stop At last layers of CLIP model": "Durdur CLIP modelinin son katmanlarında",
-    "Interrogate Options": "Sorgulama Seçenekleri",
-    "Interrogate: keep models in VRAM": "Sorgula: modelleri VRAM'de tut",
-    "Interrogate: use artists from artists.csv": "Sorgula: artists.csv dosyasındaki sanatçıları kullan",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "Interrogate: sonuçlara eşleşen model etiketlerinin sıralarını dahil et (Başlık tabanlı sorgulayıcılar üzerinde etkisi yoktur).",
-    "Interrogate: num_beams for BLIP": "Sorgula: BLIP için num_beams",
-    "Interrogate: minimum description length (excluding artists, etc..)": "Sorgula: minimum açıklama uzunluğu (sanatçılar vb. hariç)",
-    "Interrogate: maximum description length": "Sorgula: maksimum açıklama uzunluğu",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: metin dosyasındaki maksimum satır sayısı (0 = Sınır yok)",
-    "Interrogate: deepbooru score threshold": "Sorgula: deepbooru puan eşiği",
-    "Interrogate: deepbooru sort alphabetically": "Sorgula: deepbooru alfabetik olarak sırala",
-    "use spaces for tags in deepbooru": "deepbooru'da etiketler için boşluk kullanın",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "deepbooru'da kaçış (\\) parantezleri (böylece vurgu için değil, gerçek parantez olarak kullanılırlar)",
-    "User interface": "Kullanıcı arayüzü",
-    "Show progressbar": "İlerleme çubuğunu göster",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "Her N örnekleme adımında görüntü oluşturma ilerlemesini gösterir. Devre dışı bırakmak için 0 olarak ayarlayın.",
-    "Show previews of all images generated in a batch as a grid": "Bir toplu işte oluşturulan tüm görüntülerin önizlemelerini ızgara olarak göster",
-    "Show grid in results for web": "Web için sonuçlarda ızgarayı göster",
-    "Do not show any images in results for web": "Web için sonuçlarda herhangi bir resim gösterme",
-    "Add model hash to generation information": "Üretim bilgilerine model karması ekleyin",
-    "Add model name to generation information": "Üretim bilgilerine model adı ekleme",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "Üretim parametrelerini metinden kullanıcı arayüzüne okurken (PNG bilgisinden veya yapıştırılan metinden), seçilen modeli/denetim noktasını değiştirmeyin.",
-    "Font for image grids that have text": "Metin içeren görüntü ızgaraları için yazı tipi",
-    "Enable full page image viewer": "Tam sayfa resim görüntüleyiciyi etkinleştir",
-    "Show images zoomed in by default in full page image viewer": "Tam sayfa resim görüntüleyicide resimleri varsayılan olarak yakınlaştırılmış olarak gösterme",
-    "Show generation progress in window title.": "Pencere başlığında üretim ilerlemesini göster.",
-    "Quicksettings list": "Hızlı ayarlar listesi",
-    "Localization (requires restart)": "Yerelleştirme (yeniden başlatma gerektirir)",
-    "ko_KR": "ko_KR",
-    "ru_RU": "ru_RU",
-    "es_ES": "es_ES",
-    "ja_JP": "ja_JP",
-    "ar_AR": "ar_AR",
-    "Sampler parameters": "Örnekleyici parametreleri",
-    "Hide samplers in user interface (requires restart)": "Kullanıcı arayüzünde örnekleyicileri gizle (yeniden başlatma gerektirir)",
-    "eta (noise multiplier) for DDIM": "DDIM için eta (gürültü çarpanı)",
-    "eta (noise multiplier) for ancestral samplers": "eta örnekleyiciler için eta (gürültü çarpanı)",
-    "img2img DDIM discretize": "img2img DDIM discretize",
-    "uniform": "üniforma",
-    "quad": "dörtlü",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "sigma tmin",
-    "sigma noise": "sigma gürültüsü",
-    "Eta noise seed delta": "Eta gürültü tohum deltası",
-    "Request browser notifications": "Tarayıcı bildirimleri isteyin",
-    "Download localization template": "Dil şablonunu indirin",
-    "Reload custom script bodies (No ui updates, No restart)": "Kişisel komut dosyası gövdelerini yeniden yükle (Kullanıcı arayüzü güncellemesi yok, yeniden başlatma yok)",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "Gradio'yu yeniden başlatın ve bileşenleri yenileyin (yalnızca Özel Komut Dosyaları, ui.py, js ve css)",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "İstem (oluşturmak için Ctrl+Enter veya Alt+Enter tuşlarına basın)",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "Negatif istem (oluşturmak için Ctrl+Enter veya Alt+Enter tuşlarına basın)",
-    "Add a random artist to the prompt.": "Komut istemine rastgele bir sanatçı ekleyin.",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "Kullanıcı arayüzüne istemden veya istem boşsa son üretimden üretim parametrelerini okuyun.",
-    "Save style": "Stil kaydet",
-    "Apply selected styles to current prompt": "Seçilen stilleri geçerli komut istemine uygulama",
-    "Stop processing current image and continue processing.": "Geçerli görüntüyü işlemeyi durdurun ve işlemeye devam edin.",
-    "Stop processing images and return any results accumulated so far.": "Görüntüleri işlemeyi durdurun ve o ana kadar biriken tüm sonuçları döndürün.",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "Uygulanacak stil; stillerin hem pozitif hem de negatif istemler için bileşenleri vardır ve her ikisine de uygulanır",
-    "Do not do anything special": "Özel bir şey yapmayın",
-    "Which algorithm to use to produce the image": "Görüntüyü üretmek için hangi algoritmanın kullanılacağı",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - çok yaratıcı, adım sayısına bağlı olarak her biri tamamen farklı bir resim elde edebilir, adımları 30-40'tan daha yükseğe ayarlamak yardımcı olmaz",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Difüzyon Örtük Modelleri - en iyi inpainting",
-    "Produce an image that can be tiled.": "Döşenebilen bir görüntü üretin.",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "Bir görüntüyü kısmen daha düşük çözünürlükte oluşturmak, büyütmek ve ardından kompozisyonu değiştirmeden ayrıntıları iyileştirmek için iki adımlı bir işlem kullanın",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "Algoritmanın resmin içeriğine ne kadar az saygı göstermesi gerektiğini belirler. 0'da hiçbir şey değişmez ve 1'de ilgisiz bir görüntü elde edersiniz. 1,0'ın altındaki değerlerde işleme, Örnekleme Adımları kaydırıcısının belirttiğinden daha az adım atacaktır.",
-    "How many batches of images to create": "Kaç görüntü grubu oluşturulacağı",
-    "How many image to create in a single batch": "Tek bir partide kaç görüntü oluşturulacağı",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Sınıflandırıcı Serbest Rehberlik Ölçeği - görüntünün istemle ne kadar uyumlu olması gerektiği - düşük değerler daha yaratıcı sonuçlar üretir",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "Rastgele sayı üretecinin çıktısını belirleyen bir değer - başka bir resimle aynı parametrelere ve tohuma sahip bir resim oluşturursanız, aynı sonucu alırsınız",
-    "Set seed to -1, which will cause a new random number to be used every time": "Tohum değerini -1 olarak ayarlayın, bu her seferinde yeni bir rastgele sayı kullanılmasına neden olacaktır",
-    "Reuse seed from last generation, mostly useful if it was randomed": "Son nesilden tohumu yeniden kullanın, çoğunlukla rastgele ise kullanışlıdır",
-    "Seed of a different picture to be mixed into the generation.": "Nesle karıştırılacak farklı bir resmin tohumu.",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "Ne kadar güçlü bir varyasyon üretileceği. 0'da hiçbir etki olmayacaktır. 1'de, varyasyon tohumu ile tam bir resim elde edersiniz (sadece bir şey alacağınız atasal örnekleyiciler hariç).",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "Belirtilen çözünürlükte aynı tohumla üretilecek olana benzer bir resim üretme girişiminde bulunun",
-    "Separate values for X axis using commas.": "X ekseni için değerleri virgül kullanarak ayırın.",
-    "Separate values for Y axis using commas.": "Y ekseni için değerleri virgül kullanarak ayırın.",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "Görüntüyü bir dizine (varsayılan - log/images) ve üretim parametrelerini csv dosyasına yazın.",
-    "Open images output directory": "Görüntü çıktı dizinini açın",
-    "How much to blur the mask before processing, in pixels.": "İşlemeden önce maskenin piksel cinsinden ne kadar bulanıklaştırılacağı.",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "Kararlı Difüzyon ile işlemeden önce maskelenmiş alanın içine ne konulacağı.",
-    "fill it with colors of the image": "Görüntünün renkleriyle doldurun",
-    "keep whatever was there originally": "başlangıçta orada ne varsa saklayın",
-    "fill it with latent space noise": "gizli alan gürültüsü ile doldurun",
-    "fill it with latent space zeroes": "gizli uzay sıfırları ile doldurun",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "Maskelenmiş bölgeyi hedef çözünürlüğe yükseltme, inpainting yapma, ölçeği küçültme ve orijinal görüntüye yapıştırma",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "Görüntüyü hedef çözünürlüğe göre yeniden boyutlandırın. Yükseklik ve genişlik eşleşmediği sürece, yanlış en boy oranı elde edersiniz.",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "Görüntüyü, hedef çözünürlüğün tamamı görüntüyle dolacak şekilde yeniden boyutlandırın. Dışarıda kalan kısımları kırpın.",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "Görüntünün tamamı hedef çözünürlüğün içinde olacak şekilde görüntüyü yeniden boyutlandırın. Boş alanı görüntünün renkleriyle doldurun.",
-    "How many times to repeat processing an image and using it as input for the next iteration": "Bir görüntüyü işlemeyi kaç kez tekrarlamak ve bir sonraki yineleme için girdi olarak kullanmak",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "Geri döngü modunda, her döngüde denoising gücü bu değerle çarpılır. <1 çeşitliliğin azalması anlamına gelir, böylece diziniz sabit bir resme yakınsayacaktır. >1'den büyük olması çeşitliliğin artması anlamına gelir, böylece sekansınız gittikçe daha kaotik hale gelecektir.",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "SD yükseltme için karolar arasında piksel olarak ne kadar örtüşme olmalıdır. Döşemeler, tekrar tek bir resimde birleştirildiklerinde açıkça görülebilen bir dikiş olmayacak şekilde üst üste biner.",
-    "A directory on the same machine where the server is running.": "Sunucunun çalıştığı makinedeki bir dizin.",
-    "Leave blank to save images to the default path.": "Görüntüleri varsayılan yola kaydetmek için boş bırakın.",
-    "Result = A * (1 - M) + B * M": "Sonuç = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "Sonuç = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "1. ve son rakam 1 olmalıdır. örn:'1, 2, 1'",
-    "Path to directory with input images": "Girdi resimlerinin bulunduğu dizinin yolu",
-    "Path to directory where to write outputs": "Çıktıların yazılacağı dizinin yolu",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Görüntülerin dosya adlarının nasıl seçileceğini tanımlamak için aşağıdaki etiketleri kullanın: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; varsayılan için boş bırakın.",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "Bu seçenek etkinleştirilirse, oluşturulan görüntülere filigran eklenmeyecektir. Uyarı: filigran eklemezseniz, etik olmayan bir şekilde davranıyor olabilirsiniz.",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "Görüntüler ve ızgaralar için alt dizinlerin nasıl seçileceğini tanımlamak için aşağıdaki etiketleri kullanın: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; varsayılan için boş bırakın.",
-    "Restore low quality faces using GFPGAN neural network": "GFPGAN sinir ağını kullanarak düşük kaliteli yüzleri geri yükleme",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "Bu düzenli ifade, dosya adından sözcükleri ayıklamak için kullanılır ve bunlar aşağıdaki seçenek kullanılarak eğitim için kullanılan etiket metnine birleştirilir. Dosya adı metnini olduğu gibi tutmak için boş bırakın.",
-    "This string will be used to join split words into a single line if the option above is enabled.": "Bu dize, yukarıdaki seçenek etkinleştirilirse bölünmüş kelimeleri tek bir satırda birleştirmek için kullanılacaktır.",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "Normal ayar sekmesi yerine üstteki hızlı erişim çubuğuna gitmesi gereken ayarlar için virgülle ayrılmış ayar adlarının listesi. Ayar adları için modules/shared.py dosyasına bakın. Uygulanması için yeniden başlatma gerekir.",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "Bu değer sıfır değilse, tohuma eklenecek ve Eta ile örnekleyiciler kullanılırken gürültüler için RNG'yi başlatmak için kullanılacaktır. Bunu daha fazla görüntü çeşitliliği üretmek için kullanabilir veya ne yaptığınızı biliyorsanız diğer yazılımların görüntülerini eşleştirmek için kullanabilirsiniz.."
-}

localizations/zh_CN.json

@@ -1,624 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "载入中...",
-    "view": "查看",
-    "api": "api",
-    "•": " • ",
-    "built with gradio": "基于 Gradio 构建",
-    "Stable Diffusion checkpoint": "Stable Diffusion 模型(ckpt)",
-    "txt2img": "文生图",
-    "img2img": "图生图",
-    "Extras": "更多",
-    "PNG Info": "图片信息",
-    "Checkpoint Merger": "模型(ckpt)合并",
-    "Train": "训练",
-    "Create aesthetic embedding": "生成美术风格",
-    "Image Browser": "图库浏览器",
-    "Settings": "设置",
-    "Extensions": "扩展",
-    "Prompt": "提示词",
-    "Negative prompt": "反向提示词",
-    "Run": "运行",
-    "Skip": "跳过",
-    "Interrupt": "中止",
-    "Generate": "生成",
-    "Style 1": "模版风格 1",
-    "Style 2": "模版风格 2",
-    "Label": "标签",
-    "File": "文件",
-    "Drop File Here": "拖拽文件到此",
-    "-": "-",
-    "or": "或",
-    "Click to Upload": "点击上传",
-    "Image": "图像",
-    "Check progress": "查看进度",
-    "Check progress (first)": "(首次)查看进度",
-    "Sampling Steps": "采样迭代步数 (Steps)",
-    "Sampling method": "采样方法 (Sampler)",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "宽度",
-    "Height": "高度",
-    "Restore faces": "面部修复",
-    "Tiling": "可平铺(Tiling)",
-    "Highres. fix": "高分辨率修复",
-    "Firstpass width": "第一遍的宽度",
-    "Firstpass height": "第一遍的高度",
-    "Denoising strength": "重绘幅度(Denoising strength)",
-    "Batch count": "生成批次",
-    "Batch size": "每批数量",
-    "CFG Scale": "提示词相关性(CFG Scale)",
-    "Seed": "随机种子(seed)",
-    "Extra": "▼",
-    "Variation seed": "差异随机种子",
-    "Variation strength": "差异强度",
-    "Resize seed from width": "自宽度缩放随机种子",
-    "Resize seed from height": "自高度缩放随机种子",
-    "Open for Clip Aesthetic!": "打开以调整 Clip 的美术风格！",
-    "Aesthetic weight": "美术风格权重",
-    "Aesthetic steps": "美术风格迭代步数",
-    "Aesthetic learning rate": "美术风格学习率",
-    "Slerp interpolation": "球面线性插值",
-    "Aesthetic imgs embedding": "美术风格图集 embedding",
-    "None": "无",
-    "Aesthetic text for imgs": "该图集的美术风格描述",
-    "Slerp angle": "球面线性插值角度",
-    "Is negative text": "是反向提示词",
-    "Script": "脚本",
-    "Embedding to Shareable PNG": "将 Embedding 转换为可分享的 PNG 图片文件",
-    "Prompt matrix": "提示词矩阵",
-    "Prompts from file or textbox": "从文本框或文件载入提示词",
-    "X/Y plot": "X/Y 图表",
-    "Source embedding to convert": "用于转换的源 Embedding",
-    "Embedding token": "Embedding 的 token (关键词)",
-    "Output directory": "输出目录",
-    "Put variable parts at start of prompt": "把变量部分放在提示词文本的开头",
-    "Iterate seed every line": "每行输入都换一个种子",
-    "Use same random seed for all lines": "每行输入都使用同一个随机种子",
-    "List of prompt inputs": "提示词输入列表",
-    "Upload prompt inputs": "上传提示词输入文件",
-    "X type": "X轴类型",
-    "Nothing": "无",
-    "Var. seed": "差异随机种子",
-    "Var. strength": "差异强度",
-    "Steps": "迭代步数",
-    "Prompt S/R": "提示词替换",
-    "Prompt order": "提示词顺序",
-    "Sampler": "采样器",
-    "Checkpoint name": "模型(ckpt)名",
-    "Hypernetwork": "超网络(Hypernetwork)",
-    "Hypernet str.": "超网络(Hypernetwork) 强度",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "最小 Sigma",
-    "Sigma max": "最大 Sigma",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Eta",
-    "Clip skip": "Clip 跳过",
-    "Denoising": "重绘幅度",
-    "Cond. Image Mask Weight": "图像调节屏蔽度",
-    "X values": "X轴数值",
-    "Y type": "Y轴类型",
-    "Y values": "Y轴数值",
-    "Draw legend": "在图表中包括轴标题",
-    "Include Separate Images": "包括独立的图像",
-    "Keep -1 for seeds": "保持随机种子为-1",
-    "Save": "保存",
-    "Send to img2img": ">> 图生图",
-    "Send to inpaint": ">> 局部重绘",
-    "Send to extras": ">> 更多",
-    "Make Zip when Save?": "保存时生成zip压缩文件?",
-    "Textbox": "文本框",
-    "Interrogate\nCLIP": "CLIP\n反推提示词",
-    "Interrogate\nDeepBooru": "DeepBooru\n反推提示词",
-    "Inpaint": "局部重绘",
-    "Batch img2img": "批量图生图",
-    "Image for img2img": "图生图的图像",
-    "Drop Image Here": "拖拽图像到此",
-    "Image for inpainting with mask": "用于局部重绘并手动画蒙版的图像",
-    "Mask": "蒙版",
-    "Mask blur": "蒙版模糊",
-    "Mask mode": "蒙版模式",
-    "Draw mask": "绘制蒙版",
-    "Upload mask": "上传蒙版",
-    "Masking mode": "蒙版模式",
-    "Inpaint masked": "重绘蒙版内容",
-    "Inpaint not masked": "重绘非蒙版内容",
-    "Masked content": "蒙版蒙住的内容",
-    "fill": "填充",
-    "original": "原图",
-    "latent noise": "潜空间噪声",
-    "latent nothing": "潜空间数值零",
-    "Inpaint at full resolution": "全分辨率局部重绘",
-    "Inpaint at full resolution padding, pixels": "预留像素",
-    "Process images in a directory on the same machine where the server is running.": "使用服务器主机上的一个目录，作为输入目录处理图像",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "使用一个空的文件夹作为输出目录，而不是使用默认的 output 文件夹作为输出目录",
-    "Input directory": "输入目录",
-    "Resize mode": "缩放模式",
-    "Just resize": "拉伸",
-    "Crop and resize": "裁剪",
-    "Resize and fill": "填充",
-    "img2img alternative test": "图生图的另一种测试",
-    "Loopback": "回送",
-    "Outpainting mk2": "向外绘制第二版",
-    "Poor man's outpainting": "效果稍差的向外绘制",
-    "SD upscale": "使用 SD 放大(SD upscale)",
-    "should be 2 or lower.": "必须小于等于2",
-    "Override `Sampling method` to Euler?(this method is built for it)": "覆写 `采样方法` 为 Euler?(这个方法就是为这样做设计的)",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "覆写 `提示词` 为 `初始提示词`?(包括`反向提示词`)",
-    "Original prompt": "初始提示词",
-    "Original negative prompt": "初始反向提示词",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "覆写 `采样迭代步数` 为 `解码迭代步数`?",
-    "Decode steps": "解码迭代步数",
-    "Override `Denoising strength` to 1?": "覆写 `重绘幅度` 为 1?",
-    "Decode CFG scale": "解码提示词相关性(CFG scale)",
-    "Randomness": "随机度",
-    "Sigma adjustment for finding noise for image": "为寻找图中噪点的 Sigma 调整",
-    "Loops": "迭代次数",
-    "Denoising strength change factor": "重绘幅度的调整系数",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "推荐设置：采样迭代步数：80-100，采样器：Euler a，重绘幅度：0.8",
-    "Pixels to expand": "拓展的像素数",
-    "Outpainting direction": "向外绘制的方向",
-    "left": "左",
-    "right": "右",
-    "up": "上",
-    "down": "下",
-    "Fall-off exponent (lower=higher detail)": "衰减指数(越低细节越好)",
-    "Color variation": "色彩变化",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "将图像放大到两倍尺寸； 使用宽度和高度滑块设置图块尺寸(tile size)",
-    "Tile overlap": "图块重叠的像素(Tile overlap)",
-    "Upscaler": "放大算法",
-    "Lanczos": "Lanczos",
-    "Nearest": "最邻近(整数缩放)",
-    "LDSR": "LDSR",
-    "BSRGAN 4x": "BSRGAN 4x",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR_4x": "SwinIR 4x",
-    "Single Image": "单个图像",
-    "Batch Process": "批量处理",
-    "Batch from Directory": "从目录进行批量处理",
-    "Source": "来源",
-    "Show result images": "显示输出图像",
-    "Scale by": "等比缩放",
-    "Scale to": "指定尺寸缩放",
-    "Resize": "缩放",
-    "Crop to fit": "裁剪以适应",
-    "Upscaler 2 visibility": "放大算法 2 (Upscaler 2) 可见度",
-    "GFPGAN visibility": "GFPGAN 可见度",
-    "CodeFormer visibility": "CodeFormer 可见度",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormer 权重 (0 = 最大效果, 1 = 最小效果)",
-    "Upscale Before Restoring Faces": "放大后再进行面部修复",
-    "Send to txt2img": ">> 文生图",
-    "A merger of the two checkpoints will be generated in your": "合并后的模型(ckpt)会生成在你的",
-    "checkpoint": "模型(ckpt)",
-    "directory.": "目录",
-    "Primary model (A)": "主要模型 (A)",
-    "Secondary model (B)": "第二模型 (B)",
-    "Tertiary model (C)": "第三模型 (C)",
-    "Custom Name (Optional)": "自定义名称 (可选)",
-    "Multiplier (M) - set to 0 to get model A": "倍率 (M) - 设为 0 等价于模型 A",
-    "Interpolation Method": "插值方法",
-    "Weighted sum": "加权和",
-    "Add difference": "添加差分",
-    "Save as float16": "以 float16 储存",
-    "See": "查看",
-    "wiki": "wiki文档",
-    "for detailed explanation.": "以了解详细说明",
-    "Create embedding": "生成 embedding",
-    "Create hypernetwork": "生成 hypernetwork",
-    "Preprocess images": "图像预处理",
-    "Name": "名称",
-    "Initialization text": "初始化文字",
-    "Number of vectors per token": "每个 token 的向量数",
-    "Overwrite Old Embedding": "覆写旧的 Embedding",
-    "Modules": "模块",
-    "Enter hypernetwork layer structure": "输入 hypernetwork 层结构",
-    "Select activation function of hypernetwork": "选择 hypernetwork 的激活函数",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "tanh": "tanh",
-    "sigmoid": "sigmoid",
-    "celu": "celu",
-    "gelu": "gelu",
-    "glu": "glu",
-    "hardshrink": "hardshrink",
-    "hardsigmoid": "hardsigmoid",
-    "hardtanh": "hardtanh",
-    "logsigmoid": "logsigmoid",
-    "logsoftmax": "logsoftmax",
-    "mish": "mish",
-    "prelu": "prelu",
-    "rrelu": "rrelu",
-    "relu6": "relu6",
-    "selu": "selu",
-    "silu": "silu",
-    "softmax": "softmax",
-    "softmax2d": "softmax2d",
-    "softmin": "softmin",
-    "softplus": "softplus",
-    "softshrink": "softshrink",
-    "softsign": "softsign",
-    "tanhshrink": "tanhshrink",
-    "threshold": "阈值",
-    "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "选择初始化层权重的方案. 类relu - Kaiming, 类sigmoid - Xavier 都是比较推荐的选项",
-    "Normal": "正态",
-    "KaimingUniform": "Kaiming 均匀",
-    "KaimingNormal": "Kaiming 正态",
-    "XavierUniform": "Xavier 均匀",
-    "XavierNormal": "Xavier 正态",
-    "Add layer normalization": "添加层标准化",
-    "Use dropout": "采用 dropout 防止过拟合",
-    "Overwrite Old Hypernetwork": "覆写旧的 Hypernetwork",
-    "Source directory": "源目录",
-    "Destination directory": "目标目录",
-    "Existing Caption txt Action": "对已有的 txt 说明文字的行为",
-    "ignore": "无视",
-    "copy": "复制",
-    "prepend": "放前面",
-    "append": "放后面",
-    "Create flipped copies": "生成镜像副本",
-    "Split oversized images": "分割过大的图像",
-    "Auto focal point crop": "自动焦点裁切",
-    "Use BLIP for caption": "使用 BLIP 生成说明文字(自然语言描述)",
-    "Use deepbooru for caption": "使用 deepbooru 生成说明文字(tags)",
-    "Split image threshold": "图像分割阈值",
-    "Split image overlap ratio": "分割图像重叠的比率",
-    "Focal point face weight": "焦点面部权重",
-    "Focal point entropy weight": "焦点熵权重",
-    "Focal point edges weight": "焦点线条权重",
-    "Create debug image": "生成调试(debug)图片",
-    "Preprocess": "预处理",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "训练 embedding 或者 hypernetwork； 必须指定一组具有 1:1 比例图像的目录",
-    "[wiki]": "[wiki文档]",
-    "Embedding": "Embedding",
-    "Embedding Learning rate": "Embedding 学习率",
-    "Hypernetwork Learning rate": "Hypernetwork 学习率",
-    "Dataset directory": "数据集目录",
-    "Log directory": "日志目录",
-    "Prompt template file": "提示词模版文件",
-    "Max steps": "最大迭代步数",
-    "Save an image to log directory every N steps, 0 to disable": "每 N 步保存一个图像到日志目录，0 表示禁用",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "每 N 步将 embedding 的副本保存到日志目录，0 表示禁用",
-    "Save images with embedding in PNG chunks": "保存图像，并在 PNG 图片文件中嵌入 embedding 文件",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "进行预览时，从文生图选项卡中读取参数（提示词等）",
-    "Train Hypernetwork": "训练 Hypernetwork",
-    "Train Embedding": "训练 Embedding",
-    "Create an aesthetic embedding out of any number of images": "从任意数量的图像中创建美术风格 embedding",
-    "Create images embedding": "生成图集 embedding",
-    "Favorites": "收藏夹(已保存)",
-    "Others": "其他",
-    "Images directory": "图像目录",
-    "Dropdown": "下拉列表",
-    "First Page": "首页",
-    "Prev Page": "上一页",
-    "Page Index": "页数",
-    "Next Page": "下一页",
-    "End Page": "尾页",
-    "delete next": "删除下一张",
-    "Delete": "删除",
-    "sort by": "排序方式",
-    "path name": "路径名",
-    "date": "日期",
-    "keyword": "搜索",
-    "Generate Info": "生成信息",
-    "File Name": "文件名",
-    "Move to favorites": "移动到收藏夹(保存)",
-    "Renew Page": "刷新页面",
-    "Number": "数量",
-    "set_index": "设置索引",
-    "load_switch": "载入开关",
-    "turn_page_switch": "翻页开关",
-    "Checkbox": "勾选框",
-    "Apply settings": "保存设置",
-    "Saving images/grids": "保存图像/宫格图",
-    "Always save all generated images": "始终保存所有生成的图像",
-    "File format for images": "图像的文件格式",
-    "Images filename pattern": "图像文件名格式",
-    "Add number to filename when saving": "储存的时候在文件名里添加数字",
-    "Always save all generated image grids": "始终保存所有生成的宫格图",
-    "File format for grids": "宫格图的文件格式",
-    "Add extended info (seed, prompt) to filename when saving grid": "保存宫格图时，将扩展信息（随机种子、提示词）添加到文件名",
-    "Do not save grids consisting of one picture": "只有一张图片时不要保存宫格图",
-    "Prevent empty spots in grid (when set to autodetect)": "（启用自动检测时）防止宫格图中出现空位",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "宫格图行数； 使用 -1 进行自动检测，使用 0 使其与每批数量相同",
-    "Save text information about generation parameters as chunks to png files": "将有关生成参数的文本信息，作为块保存到 png 图片文件中",
-    "Create a text file next to every image with generation parameters.": "保存图像时，在每个图像旁边创建一个文本文件储存生成参数",
-    "Save a copy of image before doing face restoration.": "在进行面部修复之前保存图像副本",
-    "Save a copy of image before applying highres fix.": "在做高分辨率修复之前保存初始图像副本",
-    "Save a copy of image before applying color correction to img2img results": "在对图生图结果应用颜色校正之前保存图像副本",
-    "Quality for saved jpeg images": "保存的 jpeg 图像的质量",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "如果 PNG 图像大于 4MB 或宽高大于 4000，则缩小并保存副本为 JPG 图片",
-    "Use original name for output filename during batch process in extras tab": "在更多选项卡中的批量处理过程中，使用原始名称作为输出文件名",
-    "When using 'Save' button, only save a single selected image": "使用“保存”按钮时，只保存一个选定的图像",
-    "Do not add watermark to images": "不要给图像加水印",
-    "Paths for saving": "保存路径",
-    "Output directory for images; if empty, defaults to three directories below": "图像的输出目录； 如果为空，则默认为以下三个目录",
-    "Output directory for txt2img images": "文生图的输出目录",
-    "Output directory for img2img images": "图生图的输出目录",
-    "Output directory for images from extras tab": "更多选项卡的输出目录",
-    "Output directory for grids; if empty, defaults to two directories below": "宫格图的输出目录； 如果为空，则默认为以下两个目录",
-    "Output directory for txt2img grids": "文生图宫格的输出目录",
-    "Output directory for img2img grids": "图生图宫格的输出目录",
-    "Directory for saving images using the Save button": "使用“保存”按钮保存图像的目录",
-    "Saving to a directory": "保存到目录",
-    "Save images to a subdirectory": "将图像保存到子目录",
-    "Save grids to a subdirectory": "将宫格图保存到子目录",
-    "When using \"Save\" button, save images to a subdirectory": "使用“保存”按钮时，将图像保存到子目录",
-    "Directory name pattern": "目录名称格式",
-    "Max prompt words for [prompt_words] pattern": "[prompt_words] 格式的最大提示词数量",
-    "Upscaling": "放大",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "ESRGAN 的图块尺寸(Tile size)。0 = 不分块(no tiling)",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "ESRGAN 的图块重叠(Tile overlap)像素。低值 = 可见接缝",
-    "Tile size for all SwinIR.": "适用所有 SwinIR 系算法的图块尺寸(Tile size)",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "SwinIR 的图块重叠(Tile overlap)像素。低值 = 可见接缝",
-    "LDSR processing steps. Lower = faster": "LDSR 处理迭代步数。更低 = 更快",
-    "Upscaler for img2img": "图生图的放大算法",
-    "Upscale latent space image when doing hires. fix": "做高分辨率修复时，也放大潜空间图像",
-    "Face restoration": "面部修复",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormer 权重参数； 0 = 最大效果； 1 = 最小效果",
-    "Move face restoration model from VRAM into RAM after processing": "面部修复处理完成后，将面部修复模型从显存(VRAM)移至内存(RAM)",
-    "System": "系统",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "生成图像时，每秒轮询显存(VRAM)使用情况的次数。设置为 0 以禁用",
-    "Always print all generation info to standard output": "始终将所有生成信息输出到 standard output (一般为控制台)",
-    "Add a second progress bar to the console that shows progress for an entire job.": "向控制台添加第二个进度条，显示整个作业的进度",
-    "Training": "训练",
-    "Move VAE and CLIP to RAM when training if possible. Saves VRAM.": "训练时将 VAE 和 CLIP 从显存(VRAM)移放到内存(RAM)如果可行的话，节省显存(VRAM)",
-    "Filename word regex": "文件名用词的正则表达式",
-    "Filename join string": "文件名连接用字符串",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "每个 epoch 中单个输入图像的重复次数； 仅用于显示 epoch 数",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "每 N 步保存一个包含 loss 的 csv 表格到日志目录，0 表示禁用",
-    "Use cross attention optimizations while training": "训练时开启 cross attention 优化",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "缓存在内存(RAM)中的模型(ckpt)",
-    "SD VAE": "模型的 VAE (SD VAE)",
-    "auto": "自动",
-    "Hypernetwork strength": "Hypernetwork 强度",
-    "Inpainting conditioning mask strength": "局部重绘时图像调节的蒙版屏蔽强度",
-    "Apply color correction to img2img results to match original colors.": "对图生图结果应用颜色校正以匹配原始颜色",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "在进行图生图的时候，确切地执行滑块指定的迭代步数（正常情况下更弱的重绘幅度需要更少的迭代步数）",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "在 K 采样器中启用量化以获得更清晰、更清晰的结果。这可能会改变现有的随机种子。需要重新启动才能应用",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "强调符：使用 (文字) 使模型更关注该文本，使用 [文字] 使其减少关注",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "使用旧的强调符实现。可用于复现旧随机种子",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "使 K-diffusion 采样器 批量生成与生成单个图像时，产出相同的图像",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "当使用超过 75 个 token 时，通过从 n 个 token 中的最后一个逗号填补来提高一致性",
-    "Filter NSFW content": "过滤成人内容(NSFW)",
-    "Stop At last layers of CLIP model": "在 CLIP 模型的最后哪一层停下 (Clip skip)",
-    "Interrogate Options": "反推提示词选项",
-    "Interrogate: keep models in VRAM": "反推: 将模型保存在显存(VRAM)中",
-    "Interrogate: use artists from artists.csv": "反推: 使用 artists.csv 中的艺术家",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "反推: 在生成结果中包含与模型标签(tags)相匹配的等级（对基于生成自然语言描述的反推没有影响）",
-    "Interrogate: num_beams for BLIP": "反推: BLIP 的 num_beams",
-    "Interrogate: minimum description length (excluding artists, etc..)": "反推: 最小描述长度（不包括艺术家, 等…）",
-    "Interrogate: maximum description length": "反推: 最大描述长度",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP: 文本文件中的最大行数（0 = 无限制）",
-    "Interrogate: deepbooru score threshold": "反推: deepbooru 分数阈值",
-    "Interrogate: deepbooru sort alphabetically": "反推: deepbooru 按字母顺序排序",
-    "use spaces for tags in deepbooru": "在 deepbooru 中为标签使用空格",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "在 deepbooru 中使用转义 (\\) 括号（因此它们用作文字括号而不是强调符号）",
-    "User interface": "用户界面",
-    "Show progressbar": "显示进度条",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "每 N 个采样迭代步数显示图像生成进度。设置 0 禁用",
-    "Show previews of all images generated in a batch as a grid": "以网格的形式，预览批量生成的所有图像",
-    "Show grid in results for web": "在网页的结果中显示宫格图",
-    "Do not show any images in results for web": "不在网页的结果中显示任何图像",
-    "Add model hash to generation information": "将模型的哈希值添加到生成信息",
-    "Add model name to generation information": "将模型名称添加到生成信息",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "从文本读取生成参数到用户界面（从 PNG 图片信息或粘贴文本）时，不要更改选定的模型(ckpt)",
-    "Send seed when sending prompt or image to other interface": "将提示词或者图片发送到 >> 其他界面时，把随机种子也传送过去",
-    "Font for image grids that have text": "有文字的宫格图使用的字体",
-    "Enable full page image viewer": "启用整页图像查看器",
-    "Show images zoomed in by default in full page image viewer": "在整页图像查看器中，默认放大显示图像",
-    "Show generation progress in window title.": "在窗口标题中显示生成进度",
-    "Quicksettings list": "快速设置列表",
-    "Localization (requires restart)": "本地化翻译（需要保存设置，并重启Gradio）",
-    "Sampler parameters": "采样器参数",
-    "Hide samplers in user interface (requires restart)": "在用户界面中隐藏采样器（需要重新启动）",
-    "eta (noise multiplier) for DDIM": "DDIM 的 eta (噪声乘数) ",
-    "eta (noise multiplier) for ancestral samplers": "ancestral 采样器的 eta (噪声乘数)",
-    "img2img DDIM discretize": "图生图 DDIM 离散化",
-    "uniform": "均匀",
-    "quad": "二阶",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "最小(tmin) sigma",
-    "sigma noise": "sigma 噪声",
-    "Eta noise seed delta": "Eta 噪声种子偏移(ENSD - Eta noise seed delta)",
-    "Images Browser": "图库浏览器",
-    "Preload images at startup": "在启动时预加载图像",
-    "Number of columns on the page": "每页列数",
-    "Number of rows on the page": "每页行数",
-    "Minimum number of pages per load": "每次加载的最小页数",
-    "Use same seed for all images": "为所有图像使用同一个随机种子",
-    "Request browser notifications": "请求浏览器通知",
-    "Download localization template": "下载本地化模板",
-    "Reload custom script bodies (No ui updates, No restart)": "重新加载自定义脚本主体（无用户界面更新，无重启）",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "重启 Gradio 及刷新组件（仅限自定义脚本、ui.py、js 和 css）",
-    "Available": "可用",
-    "Install from URL": "从网址安装",
-    "Apply and restart UI": "应用并重启用户界面",
-    "Check for updates": "检查更新",
-    "Extension": "扩展",
-    "URL": "网址",
-    "Update": "更新",
-    "a1111-sd-webui-tagcomplete": "Tag自动补全",
-    "unknown": "未知",
-    "deforum-for-automatic1111-webui": "Deforum",
-    "sd-dynamic-prompting": "动态提示词",
-    "stable-diffusion-webui-aesthetic-gradients": "美术风格梯度",
-    "stable-diffusion-webui-aesthetic-image-scorer": "美术风格评分",
-    "stable-diffusion-webui-artists-to-study": "艺术家图库",
-    "stable-diffusion-webui-dataset-tag-editor": "数据集标签编辑器",
-    "stable-diffusion-webui-images-browser": "图库浏览器",
-    "stable-diffusion-webui-inspiration": "灵感",
-    "stable-diffusion-webui-wildcards": "通配符",
-    "Load from:": "加载自",
-    "Extension index URL": "扩展列表链接",
-    "URL for extension's git repository": "扩展的 git 仓库链接",
-    "Local directory name": "本地路径名",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "提示词（按 Ctrl+Enter 或 Alt+Enter 生成）\nPrompt",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "反向提示词（按 Ctrl+Enter 或 Alt+Enter 生成）\nNegative prompt",
-    "Stop processing current image and continue processing.": "停止处理当前图像，并继续处理下一个",
-    "Stop processing images and return any results accumulated so far.": "停止处理图像，并返回迄今为止累积的任何结果",
-    "Do not do anything special": "什么都不做",
-    "Which algorithm to use to produce the image": "使用哪种算法生成图像",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - 非常有创意，可以根据迭代步数获得完全不同的图像，将迭代步数设置为高于 30-40 不会有正面作用",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit models - 最擅长局部重绘",
-    "Produce an image that can be tiled.": "生成可用于平铺(tiled)的图像",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "使用两步处理的时候，以较小的分辨率生成初步图像、接着放大图像，然后在不更改构图的情况下改进其中的细节",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "决定算法对图像内容的影响程度。设置 0 时，什么都不会改变，而在 1 时，你将获得不相关的图像。\n值低于 1.0 时，处理的迭代步数将少于“采样迭代步数”滑块指定的步数",
-    "How many batches of images to create": "创建多少批次的图像",
-    "How many image to create in a single batch": "每批创建多少图像",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - 图像应在多大程度上服从提示词 - 较低的值会产生更有创意的结果",
-    "Seed of a different picture to be mixed into the generation.": "将要参与生成的另一张图的随机种子",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "想要产生多强烈的变化。设为 0 时，将没有效果。设为 1 时，你将获得完全产自差异随机种子的图像（ancestral 采样器除外，你只是单纯地生成了一些东西）",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "尝试生成与在指定分辨率下使用相同随机种子生成的图像相似的图片",
-    "This text is used to rotate the feature space of the imgs embs": "此文本用于旋转图集 embeddings 的特征空间",
-    "Separate values for X axis using commas.": "使用逗号分隔 X 轴的值",
-    "Separate values for Y axis using commas.": "使用逗号分隔 Y 轴的值",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "将图像写入目录（默认 - log/images）并将生成参数写入 csv 表格文件",
-    "How much to blur the mask before processing, in pixels.": "处理前要对蒙版进行多强的模糊，以像素为单位",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "在使用 Stable Diffusion 处理蒙版区域之前要在蒙版区域内放置什么",
-    "fill it with colors of the image": "用图像的颜色(高强度模糊)填充它",
-    "keep whatever was there originally": "保留原来的图像，不进行预处理",
-    "fill it with latent space noise": "于潜空间填充噪声",
-    "fill it with latent space zeroes": "于潜空间填零",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "将蒙版区域（包括预留像素长度的缓冲区域）放大到目标分辨率，进行局部重绘。\n然后缩小并粘贴回原始图像中",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "将图像大小调整为目标分辨率。除非高度和宽度匹配，否则你将获得不正确的纵横比",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "调整图像大小，使整个目标分辨率都被图像填充。裁剪多出来的部分",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "调整图像大小，使整个图像在目标分辨率内。用图像的颜色填充空白区域",
-    "How many times to repeat processing an image and using it as input for the next iteration": "重复处理图像并用作下次迭代输入的次数",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "在回送模式下，在每个循环中，重绘幅度都会乘以该值。<1 表示减少多样性，因此你的这一组图将集中在固定的图像上。>1 意味着增加多样性，因此你的这一组图将变得越来越混乱",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "使用 SD 放大(SD upscale)时，图块(Tiles)之间应该有多少像素重叠。图块(Tiles)之间需要重叠才可以让它们在合并回一张图像时，没有清晰可见的接缝",
-    "A directory on the same machine where the server is running.": "与服务器主机上的目录",
-    "Leave blank to save images to the default path.": "留空以将图像保存到默认路径",
-    "Result = A * (1 - M) + B * M": "结果 = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "结果 = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "第一个和最后一个数字必须是 1。例:'1, 2, 1'",
-    "Path to directory with input images": "带有输入图像的路径",
-    "Path to directory where to write outputs": "进行输出的路径",
-    "Input images directory": "输入图像目录",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "使用以下标签定义如何选择图像的文件名: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; 默认请留空",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "如果启用此选项，水印将不会添加到生成出来的图像中。警告：如果你不添加水印，你的行为可能是不符合专业操守的",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "使用以下标签定义如何选择图像和宫格图的子目录: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; 默认请留空",
-    "Restore low quality faces using GFPGAN neural network": "使用 GFPGAN 神经网络修复低质量面部",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "此正则表达式将用于从文件名中提取单词，并将使用以下选项将它们接合到用于训练的标签文本中。留空以保持文件名文本不变",
-    "This string will be used to join split words into a single line if the option above is enabled.": "如果启用了上述选项，则此处的字符会用于将拆分的单词接合为同一行",
-    "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "仅适用于局部重绘专用的模型（模型后缀为 inpainting.ckpt 的模型）。决定了蒙版在局部重绘以及图生图中屏蔽原图内容的强度。 1.0 表示完全屏蔽原图，这是默认行为。0.0 表示完全不屏蔽让原图进行图像调节。较低的值将有助于保持原图的整体构图，但很难遇到较大的变化",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "设置项名称的列表，以逗号分隔，该设置会移动到顶部的快速访问栏，而不是默认的设置选项卡。有关设置名称，请参见 modules/shared.py。需要重新启动才能应用",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "如果这个值不为零，它将被添加到随机种子中，并在使用带有 Eta 的采样器时用于初始化随机噪声。你可以使用它来产生更多的图像变化，或者你可以使用它来模仿其他软件生成的图像，如果你知道你在做什么",
-    "Leave empty for auto": "留空时自动生成",
-    "Autocomplete options": "自动补全选项",
-    "Enable Autocomplete": "开启Tag补全",
-    "Append commas": "附加逗号",
-    "latest": "最新",
-    "behind": "落后",
-    "Roll three": "抽三位出来",
-    "Generate forever": "无限生成",
-    "Cancel generate forever": "停止无限生成",
-    "how fast should the training go. Low values will take longer to train, high values may fail to converge (not generate accurate results) and/or may break the embedding (This has happened if you see Loss: nan in the training info textbox. If this happens, you need to manually restore your embedding from an older not-broken backup).\n\nYou can set a single numeric value, or multiple learning rates using the syntax:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nWill train with rate of 0.005 for first 100 steps, then 1e-3 until 1000 steps, then 1e-5 for all remaining steps.": "训练应该多快。低值将需要更长的时间来训练，高值可能无法收敛（无法产生准确的结果）以及/也许可能会破坏 embedding（如果你在训练信息文本框中看到 Loss: nan 就会发生这种情况。如果发生这种情况，你需要从较旧的未损坏的备份手动恢复 embedding）\n\n你可以使用以下语法设置单个数值或多个学习率：\n\n   率1:步限1, 率2:步限2, ...\n\n如:   0.005:100, 1e-3:1000, 1e-5\n\n即前 100 步将以 0.005 的速率训练，接着直到 1000 步为止以 1e-3 训练，然后剩余所有步以 1e-5 训练",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "训练时将 VAE 和 CLIP 从显存(VRAM)移放到内存(RAM)，节省显存(VRAM)",
-    "How many times to improve the generated image iteratively; higher values take longer; very low values can produce bad results": "迭代改进生成的图像多少次；更高的值需要更长的时间；非常低的值会产生不好的结果",
-    "Draw a mask over an image, and the script will regenerate the masked area with content according to prompt": "在图像上画一个蒙版，脚本会根据提示重新生成蒙版区域的内容",
-    "Upscale image normally, split result into tiles, improve each tile using img2img, merge whole image back": "正常放大图像，将结果分割成图块（tiles），用图生图改进每个图块（tiles），最后将整个图像合并回来",
-    "Create a grid where images will have different parameters. Use inputs below to specify which parameters will be shared by columns and rows": "创建一个网格，图像将有不同的参数。使用下面的输入来指定哪些参数将由列和行共享",
-    "Run Python code. Advanced user only. Must run program with --allow-code for this to work": "运行 Python 代码。仅限老手使用。必须以 --allow-code 来开启程序，才能使其运行",
-    "Separate a list of words with commas, and the first word will be used as a keyword: script will search for this word in the prompt, and replace it with others": "以逗号分割的单词列表，第一个单词将被用作关键词：脚本将在提示词中搜索这个单词，并用其他单词替换它",
-    "Separate a list of words with commas, and the script will make a variation of prompt with those words for their every possible order": "以逗号分割的单词列表，脚本会排列出这些单词的所有排列方式，并加入提示词各生成一次",
-    "Reconstruct prompt from existing image and put it into the prompt field.": "从现有的图像中重构出提示词，并将其放入提示词的输入文本框",
-    "Set the maximum number of words to be used in the [prompt_words] option; ATTENTION: If the words are too long, they may exceed the maximum length of the file path that the system can handle": "设置在[prompt_words]选项中要使用的最大字数；注意：如果字数太长，可能会超过系统可处理的文件路径的最大长度",
-    "Process an image, use it as an input, repeat.": "处理一张图像，将其作为输入，并重复",
-    "Insert selected styles into prompt fields": "在提示词中插入选定的模版风格",
-    "Save current prompts as a style. If you add the token {prompt} to the text, the style use that as placeholder for your prompt when you use the style in the future.": "将当前的提示词保存为模版风格。如果你在文本中添加{prompt}标记，那么将来你使用该模版风格时，你现有的提示词会替换模版风格中的{prompt}",
-    "Loads weights from checkpoint before making images. You can either use hash or a part of filename (as seen in settings) for checkpoint name. Recommended to use with Y axis for less switching.": "在生成图像之前从模型(ckpt)中加载权重。你可以使用哈希值或文件名的一部分（如设置中所示）作为模型(ckpt)名称。建议用在Y轴上以减少过程中模型的切换",
-    "Torch active: Peak amount of VRAM used by Torch during generation, excluding cached data.\nTorch reserved: Peak amount of VRAM allocated by Torch, including all active and cached data.\nSys VRAM: Peak amount of VRAM allocation across all applications / total GPU VRAM (peak utilization%).": "Torch active: 在生成过程中，Torch使用的显存(VRAM)峰值，不包括缓存的数据。\nTorch reserved: Torch 分配的显存(VRAM)的峰值量，包括所有活动和缓存数据。\nSys VRAM: 所有应用程序分配的显存(VRAM)的峰值量 / GPU 的总显存(VRAM)（峰值利用率%）",
-    "Uscale the image in latent space. Alternative is to produce the full image from latent representation, upscale that, and then move it back to latent space.": "放大潜空间中的图像。而另一种方法是，从潜变量表达中直接解码并生成完整的图像，接着放大它，然后再将其编码回潜空间",
-    "Start drawing": "开始绘制",
-    "Description": "描述",
-    "Action": "行动",
-    "Aesthetic Gradients": "美术风格梯度",
-    "aesthetic-gradients": "美术风格梯度",
-    "Wildcards": "通配符",
-    "Dynamic Prompts": "动态提示词",
-    "Image browser": "图库浏览器",
-    "images-browser": "图库浏览器",
-    "Inspiration": "灵感",
-    "Deforum": "Deforum",
-    "Artists to study": "艺术家图库",
-    "Aesthetic Image Scorer": "美术风格评分",
-    "Dataset Tag Editor": "数据集标签编辑器",
-
-
-    "----not work----": "----以下内容无法被翻译，Bug----",
-    "Add a random artist to the prompt.": "随机添加一个艺术家到提示词中",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "从提示词中读取生成参数，如果提示词为空，则读取上一次的生成参数到用户界面",
-    "Save style": "储存为模版风格",
-    "Apply selected styles to current prompt": "将所选模板风格，应用于当前提示词",
-    "Set seed to -1, which will cause a new random number to be used every time": "将随机种子设置为-1，则每次都会使用一个新的随机数",
-    "Reuse seed from last generation, mostly useful if it was randomed": "重用上一次使用的随机种子，如果想要固定结果就会很有用",
-    "Open images output directory": "打开图像输出目录",
-    "Upscaler 1": "放大算法 1",
-    "Upscaler 2": "放大算法 2",
-    "Separate prompts into parts using vertical pipe character (|) and the script will create a picture for every combination of them (except for the first part, which will be present in all combinations)": "用竖线分隔符(|)将提示词分成若干部分，脚本将为它们的每一个组合创建一幅图片（除了被分割的第一部分，所有的组合都会包含这部分）",
-    "Select which Real-ESRGAN models to show in the web UI. (Requires restart)": "选择哪些 Real-ESRGAN 模型显示在网页用户界面。(需要重新启动)",
-    "Allowed categories for random artists selection when using the Roll button": "使用抽选艺术家按钮时将会随机的艺术家类别",
-    "Face restoration model": "面部修复模型",
-    "Install": "安装",
-    "Installing...": "安装中...",
-    "Installed": "已安装",
-
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "要使用的模版风格； 模版风格包含正向和反向提示词，并应用于两者\n\ud83c\udfa8    随机添加一个艺术家到提示词中\n \u2199\ufe0f     从提示词中读取生成参数，如果提示词为空，则读取上一次的生成参数到用户界面\n\ud83d\udcbe    将当前的提示词保存为模版风格(保存在styles.csv)\n\ud83d\udccb    将所选模板风格，应用于当前提示词\n如果你在文本中添加{prompt}标记，并保存为模版风格\n那么将来你使用该模版风格时，你现有的提示词会替换模版风格中的{prompt}",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "一个固定随机数生成器输出的值 - 以相同参数和随机种子生成的图像会得到相同的结果\n\ud83c\udfb2    将随机种子设置为-1，则每次都会使用一个新的随机数\n\u267b\ufe0f    重用上一次使用的随机种子，如果想要固定输出结果就会很有用",
-
-
-    "----deprecated----": "----以下内容在webui新版本已移除----",
-    "▼": "▼",
-    "History": "历史记录",
-    "Show Textbox": "显示文本框",
-    "File with inputs": "含输入内容的文件",
-    "Prompts": "提示词",
-    "Disabled when launched with --hide-ui-dir-config.": "启动 --hide-ui-dir-config 时禁用",
-    "Open output directory": "打开输出目录",
-    "Create aesthetic images embedding": "生成美术风格图集 embedding",
-    "Split oversized images into two": "将过大的图像分为两份",
-    "Train an embedding; must specify a directory with a set of 1:1 ratio images": "训练 embedding； 必须指定一组具有 1:1 比例图像的目录",
-    "Learning rate": "学习率",
-    "txt2img history": "文生图历史记录",
-    "img2img history": "图生图历史记录",
-    "extras history": "更多选项卡的历史记录",
-    "extras": "更多",
-    "custom fold": "自定义文件夹",
-    "Load": "载入",
-    "Prev batch": "上一批",
-    "Next batch": "下一批",
-    "number of images to delete consecutively next": "接下来要连续删除的图像数",
-    "Date to": "日期至",
-    "Refresh page": "刷新页面",
-    "Unload VAE and CLIP from VRAM when training": "训练时从显存(VRAM)中取消 VAE 和 CLIP 的加载",
-    "Number of pictures displayed on each page": "每页显示的图像数量",
-    "Number of grids in each row": "每行显示多少格",
-    "favorites": "收藏夹(已保存)",
-    "others": "其他",
-    "Collect": "收藏(保存)",
-
-
-    "--------": "--------"
-}

localizations/zh_TW.json

@@ -1,610 +0,0 @@
-{
-    "⤡": "⤡",
-    "⊞": "⊞",
-    "×": "×",
-    "❮": "❮",
-    "❯": "❯",
-    "Loading...": "載入中…",
-    "view": "檢視",
-    "api": "api",
-    "•": " • ",
-    "built with gradio": "基於 Gradio 構建",
-    "Stable Diffusion checkpoint": "Stable Diffusion 模型權重存檔點",
-    "txt2img": "文生圖",
-    "img2img": "圖生圖",
-    "Extras": "更多",
-    "PNG Info": "圖片資訊",
-    "Checkpoint Merger": "模型權重存檔點合併",
-    "Train": "訓練",
-    "Create aesthetic embedding": "生成美術風格",
-    "Image Browser": "圖庫瀏覽器",
-    "History": "歷史記錄",
-    "Settings": "設定",
-    "Extensions": "擴充",
-    "Prompt": "提示詞",
-    "Negative prompt": "反向提示詞",
-    "Run": "執行",
-    "Skip": "跳過",
-    "Interrupt": "中止",
-    "Generate": "生成",
-    "Style 1": "模版風格 1",
-    "Style 2": "模版風格 2",
-    "Label": "標記",
-    "File": "檔案",
-    "Drop File Here": "拖曳檔案到此",
-    "-": "-",
-    "or": "或",
-    "Click to Upload": "點擊上傳",
-    "Image": "圖像",
-    "Check progress": "檢視進度",
-    "Check progress (first)": "（首次）檢視進度",
-    "Sampling Steps": "採樣疊代步數",
-    "Sampling method": "採樣方法",
-    "Euler a": "Euler a",
-    "Euler": "Euler",
-    "LMS": "LMS",
-    "Heun": "Heun",
-    "DPM2": "DPM2",
-    "DPM2 a": "DPM2 a",
-    "DPM fast": "DPM fast",
-    "DPM adaptive": "DPM adaptive",
-    "LMS Karras": "LMS Karras",
-    "DPM2 Karras": "DPM2 Karras",
-    "DPM2 a Karras": "DPM2 a Karras",
-    "DDIM": "DDIM",
-    "PLMS": "PLMS",
-    "Width": "寬度",
-    "Height": "高度",
-    "Restore faces": "面部修復",
-    "Tiling": "可平鋪",
-    "Highres. fix": "高解析度修復",
-    "Firstpass width": "第一遍的寬度",
-    "Firstpass height": "第一遍的高度",
-    "Denoising strength": "重繪幅度",
-    "Batch count": "生成批次",
-    "Batch size": "每批數量",
-    "CFG Scale": "提示詞相關性（CFG）",
-    "Seed": "隨機種子",
-    "Extra": "▼",
-    "Variation seed": "差異隨機種子",
-    "Variation strength": "差異強度",
-    "Resize seed from width": "自寬度縮放隨機種子",
-    "Resize seed from height": "自高度縮放隨機種子",
-    "Open for Clip Aesthetic!": "打開以調整 Clip 的美術風格！",
-    "▼": "▼",
-    "Aesthetic weight": "美術風格權重",
-    "Aesthetic steps": "美術風格疊代步數",
-    "Aesthetic learning rate": "美術風格學習率",
-    "Slerp interpolation": "球面線性插值角度",
-    "Aesthetic imgs embedding": "美術風格圖集 embedding",
-    "None": "無",
-    "Aesthetic text for imgs": "該圖集的美術風格描述",
-    "Slerp angle": "Slerp 角度",
-    "Is negative text": "是反向提示詞",
-    "Script": "指令碼",
-    "Embedding to Shareable PNG": "將 Embedding 轉換為可分享的 PNG 圖片檔案",
-    "Prompt matrix": "提示詞矩陣",
-    "Prompts from file or textbox": "從文字方塊或檔案載入提示詞",
-    "X/Y plot": "X/Y 圖表",
-    "Source embedding to convert": "用於轉換的源 Embedding",
-    "Embedding token": "Embedding 的關鍵詞",
-    "Put variable parts at start of prompt": "把變量部分放在提示詞文本的開頭",
-    "Show Textbox": "顯示文字方塊",
-    "File with inputs": "含輸入內容的檔案",
-    "Prompts": "提示詞",
-    "Iterate seed every line": "每行輸入都換一個種子",
-    "Use same random seed for all lines": "每行輸入都使用同一個隨機種子",
-    "List of prompt inputs": "提示詞輸入列表",
-    "Upload prompt inputs": "上傳提示詞輸入檔案",
-    "X type": "X軸類型",
-    "Nothing": "無",
-    "Var. seed": "差異隨機種子",
-    "Var. strength": "差異強度",
-    "Steps": "疊代步數",
-    "Prompt S/R": "提示詞替換",
-    "Prompt order": "提示詞順序",
-    "Sampler": "採樣器",
-    "Checkpoint name": "模型權重存檔點的名稱",
-    "Hypernetwork": "超網路（Hypernetwork）",
-    "Hypernet str.": "超網路（Hypernetwork）強度",
-    "Sigma Churn": "Sigma Churn",
-    "Sigma min": "最小 Sigma",
-    "Sigma max": "最大 Sigma",
-    "Sigma noise": "Sigma noise",
-    "Eta": "Eta",
-    "Clip skip": "Clip 跳過",
-    "Denoising": "重繪幅度",
-    "Cond. Image Mask Weight": "圖像調節屏蔽度",
-    "X values": "X軸數值",
-    "Y type": "Y軸類型",
-    "Y values": "Y軸數值",
-    "Draw legend": "在圖表中包括軸標題",
-    "Include Separate Images": "包括獨立的圖像",
-    "Keep -1 for seeds": "保持隨機種子為-1",
-    "Drop Image Here": "拖曳圖像到此",
-    "Save": "儲存",
-    "Send to img2img": ">> 圖生圖",
-    "Send to inpaint": ">> 局部重繪",
-    "Send to extras": ">> 更多",
-    "Make Zip when Save?": "儲存時生成ZIP壓縮檔案？",
-    "Textbox": "文字方塊",
-    "Interrogate\nCLIP": "CLIP\n反推提示詞",
-    "Interrogate\nDeepBooru": "DeepBooru\n反推提示詞",
-    "Inpaint": "局部重繪",
-    "Batch img2img": "批量圖生圖",
-    "Image for img2img": "圖生圖的圖像",
-    "Image for inpainting with mask": "用於局部重繪並手動畫蒙版的圖像",
-    "Mask": "蒙版",
-    "Mask blur": "蒙版模糊",
-    "Mask mode": "蒙版模式",
-    "Draw mask": "繪製蒙版",
-    "Upload mask": "上傳蒙版",
-    "Masking mode": "蒙版模式",
-    "Inpaint masked": "重繪蒙版內容",
-    "Inpaint not masked": "重繪非蒙版內容",
-    "Masked content": "蒙版蒙住的內容",
-    "fill": "填充",
-    "original": "原圖",
-    "latent noise": "潛空間噪聲",
-    "latent nothing": "潛空間數值零",
-    "Inpaint at full resolution": "全解析度局部重繪",
-    "Inpaint at full resolution padding, pixels": "預留畫素",
-    "Process images in a directory on the same machine where the server is running.": "使用伺服器主機上的一個目錄，作為輸入目錄處理圖像",
-    "Use an empty output directory to save pictures normally instead of writing to the output directory.": "使用一個空的資料夾作為輸出目錄，而不是使用預設的 output 資料夾作為輸出目錄",
-    "Disabled when launched with --hide-ui-dir-config.": "啟動 --hide-ui-dir-config 時禁用",
-    "Input directory": "輸入目錄",
-    "Output directory": "輸出目錄",
-    "Resize mode": "縮放模式",
-    "Just resize": "拉伸",
-    "Crop and resize": "裁剪",
-    "Resize and fill": "填充",
-    "img2img alternative test": "圖生圖的另一種測試",
-    "Loopback": "回送",
-    "Outpainting mk2": "向外繪製第二版",
-    "Poor man's outpainting": "效果稍差的向外繪製",
-    "SD upscale": "使用 SD 放大",
-    "should be 2 or lower.": "必須小於等於2",
-    "Override `Sampling method` to Euler?(this method is built for it)": "覆寫「採樣方法」為 Euler？（這個方法就是為這樣做設計的）",
-    "Override `prompt` to the same value as `original prompt`?(and `negative prompt`)": "覆寫「提示詞」為「初始提示詞」？（包括「反向提示詞」）",
-    "Original prompt": "初始提示詞",
-    "Original negative prompt": "初始反向提示詞",
-    "Override `Sampling Steps` to the same value as `Decode steps`?": "覆寫「採樣疊代步數」為「解碼疊代步數」？",
-    "Decode steps": "解碼疊代步數",
-    "Override `Denoising strength` to 1?": "覆寫「重繪幅度」為1?",
-    "Decode CFG scale": "解碼提示詞相關性（CFG）",
-    "Randomness": "隨機度",
-    "Sigma adjustment for finding noise for image": "為尋找圖中噪點的 Sigma 調整",
-    "Loops": "疊代次數",
-    "Denoising strength change factor": "重繪幅度的調整係數",
-    "Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8": "推薦設定：採樣疊代步數：80-100，採樣器：Euler a，重繪幅度：0.8",
-    "Pixels to expand": "拓展的畫素數",
-    "Outpainting direction": "向外繪製的方向",
-    "left": "左",
-    "right": "右",
-    "up": "上",
-    "down": "下",
-    "Fall-off exponent (lower=higher detail)": "衰減指數（越低細節越好）",
-    "Color variation": "色彩變化",
-    "Will upscale the image to twice the dimensions; use width and height sliders to set tile size": "將圖像放大到兩倍尺寸； 使用寬度和高度滑塊設定圖塊尺寸",
-    "Tile overlap": "圖塊重疊的畫素",
-    "Upscaler": "放大演算法",
-    "Lanczos": "Lanczos",
-    "Nearest": "最鄰近（整數縮放）",
-    "LDSR": "LDSR",
-    "BSRGAN 4x": "BSRGAN 4x",
-    "ESRGAN_4x": "ESRGAN_4x",
-    "R-ESRGAN 4x+ Anime6B": "R-ESRGAN 4x+ Anime6B",
-    "ScuNET GAN": "ScuNET GAN",
-    "ScuNET PSNR": "ScuNET PSNR",
-    "SwinIR_4x": "SwinIR 4x",
-    "Single Image": "單個圖像",
-    "Batch Process": "批量處理",
-    "Batch from Directory": "從目錄進行批量處理",
-    "Source": "來源",
-    "Show result images": "顯示輸出圖像",
-    "Scale by": "等比縮放",
-    "Scale to": "指定尺寸縮放",
-    "Resize": "縮放",
-    "Crop to fit": "裁剪以適應",
-    "Upscaler 2": "放大演算法 2",
-    "Upscaler 2 visibility": "放大演算法 2 可見度",
-    "GFPGAN visibility": "GFPGAN 可見度",
-    "CodeFormer visibility": "CodeFormer 可見度",
-    "CodeFormer weight (0 = maximum effect, 1 = minimum effect)": "CodeFormer 權重 （0 = 最大效果， 1 = 最小效果）",
-    "Open output directory": "打開輸出目錄",
-    "Upscale Before Restoring Faces": "放大後再進行面部修復",
-    "Send to txt2img": ">> 文生圖",
-    "A merger of the two checkpoints will be generated in your": "合併後的模型權重存檔點會生成在你的",
-    "checkpoint": "模型權重存檔點",
-    "directory.": "目錄",
-    "Primary model (A)": "主要模型 (A)",
-    "Secondary model (B)": "第二模型 (B)",
-    "Tertiary model (C)": "第三模型 (C)",
-    "Custom Name (Optional)": "自訂名稱 （可選）",
-    "Multiplier (M) - set to 0 to get model A": "倍率 (M) - 設為 0 等價於模型 A",
-    "Interpolation Method": "插值方法",
-    "Weighted sum": "加權和",
-    "Add difference": "加入差分",
-    "Save as float16": "以 float16 儲存",
-    "See": "檢視",
-    "wiki": "wiki文件",
-    "for detailed explanation.": "以了解詳細說明",
-    "Create embedding": "生成 embedding",
-    "Create aesthetic images embedding": "生成美術風格圖集 embedding",
-    "Create hypernetwork": "生成超網路（Hypernetwork）",
-    "Preprocess images": "圖像預處理",
-    "Name": "名稱",
-    "Initialization text": "初始化文字",
-    "Number of vectors per token": "每個 token 的向量數",
-    "Overwrite Old Embedding": "覆寫舊的 Embedding",
-    "Modules": "模組",
-    "Enter hypernetwork layer structure": "輸入超網路（Hypernetwork）層結構",
-    "Select activation function of hypernetwork": "選擇超網路（Hypernetwork）的激活函數",
-    "linear": "linear",
-    "relu": "relu",
-    "leakyrelu": "leakyrelu",
-    "elu": "elu",
-    "swish": "swish",
-    "tanh": "tanh",
-    "sigmoid": "sigmoid",
-    "celu": "celu",
-    "gelu": "gelu",
-    "glu": "glu",
-    "hardshrink": "hardshrink",
-    "hardsigmoid": "hardsigmoid",
-    "hardtanh": "hardtanh",
-    "logsigmoid": "logsigmoid",
-    "logsoftmax": "logsoftmax",
-    "mish": "mish",
-    "prelu": "prelu",
-    "rrelu": "rrelu",
-    "relu6": "relu6",
-    "selu": "selu",
-    "silu": "silu",
-    "softmax": "softmax",
-    "softmax2d": "softmax2d",
-    "softmin": "softmin",
-    "softplus": "softplus",
-    "softshrink": "softshrink",
-    "softsign": "softsign",
-    "tanhshrink": "tanhshrink",
-    "threshold": "閾值",
-    "Select Layer weights initialization. relu-like - Kaiming, sigmoid-like - Xavier is recommended": "挑選初始化層權重的方案. 類relu - Kaiming, 類sigmoid - Xavier 都是比較推薦的選項",
-    "Normal": "正態",
-    "KaimingUniform": "Kaiming 均勻",
-    "KaimingNormal": "Kaiming 正態",
-    "XavierUniform": "Xavier 均勻",
-    "XavierNormal": "Xavier 正態",
-    "Add layer normalization": "加入層標準化",
-    "Use dropout": "採用 dropout 防止過擬合",
-    "Overwrite Old Hypernetwork": "覆寫舊的超網路（Hypernetwork）",
-    "Source directory": "來源目錄",
-    "Destination directory": "目標目錄",
-    "Existing Caption txt Action": "對已有的TXT說明文字的行為",
-    "ignore": "無視",
-    "copy": "複製",
-    "prepend": "放前面",
-    "append": "放後面",
-    "Create flipped copies": "生成鏡像副本",
-    "Split oversized images into two": "將過大的圖像分為兩份",
-    "Split oversized images": "分割過大的圖像",
-    "Auto focal point crop": "自動焦點裁切",
-    "Use BLIP for caption": "使用 BLIP 生成說明文字（自然語言描述）",
-    "Use deepbooru for caption": "使用 deepbooru 生成說明文字（標記）",
-    "Split image threshold": "圖像分割閾值",
-    "Split image overlap ratio": "分割圖像重疊的比率",
-    "Focal point face weight": "焦點面部權重",
-    "Focal point entropy weight": "焦點熵權重",
-    "Focal point edges weight": "焦點線條權重",
-    "Create debug image": "生成除錯圖片",
-    "Preprocess": "預處理",
-    "Train an embedding; must specify a directory with a set of 1:1 ratio images": "訓練 embedding； 必須指定一組具有 1:1 比例圖像的目錄",
-    "Train an embedding or Hypernetwork; you must specify a directory with a set of 1:1 ratio images": "訓練 embedding 或者超網路（Hypernetwork）； 必須指定一組具有 1:1 比例圖像的目錄",
-    "[wiki]": "[wiki文件]",
-    "Embedding": "Embedding",
-    "Embedding Learning rate": "Embedding 學習率",
-    "Hypernetwork Learning rate": "超網路（Hypernetwork）學習率",
-    "Learning rate": "學習率",
-    "Dataset directory": "資料集目錄",
-    "Log directory": "日誌目錄",
-    "Prompt template file": "提示詞模版檔案",
-    "Max steps": "最大疊代步數",
-    "Save an image to log directory every N steps, 0 to disable": "每 N 步儲存一個圖像到日誌目錄，0 表示禁用",
-    "Save a copy of embedding to log directory every N steps, 0 to disable": "每 N 步將 embedding 的副本儲存到日誌目錄，0 表示禁用",
-    "Save images with embedding in PNG chunks": "儲存圖像，並在 PNG 圖片檔案中嵌入 embedding 檔案",
-    "Read parameters (prompt, etc...) from txt2img tab when making previews": "進行預覽時，從文生圖頁籤中讀取參數（提示詞等）",
-    "Train Hypernetwork": "訓練超網路（Hypernetwork）",
-    "Train Embedding": "訓練 Embedding",
-    "Create an aesthetic embedding out of any number of images": "從任意數量的圖像中建立美術風格 embedding",
-    "Create images embedding": "生成圖集 embedding",
-    "txt2img history": "文生圖歷史記錄",
-    "img2img history": "圖生圖歷史記錄",
-    "extras history": "後處理歷史記錄",
-    "extras": "後處理",
-    "favorites": "收藏夾",
-    "Favorites": "收藏夾",
-    "Others": "其他",
-    "custom fold": "自訂資料夾",
-    "Load": "載入",
-    "Images directory": "圖像目錄",
-    "Prev batch": "上一批",
-    "Next batch": "下一批",
-    "Dropdown": "下拉式清單",
-    "First Page": "首頁",
-    "Prev Page": "上一頁",
-    "Page Index": "頁數",
-    "Next Page": "下一頁",
-    "End Page": "尾頁",
-    "number of images to delete consecutively next": "接下來要連續刪除的圖像數",
-    "delete next": "刪除下一張",
-    "Delete": "刪除",
-    "sort by": "排序方式",
-    "path name": "路徑名",
-    "date": "日期",
-    "keyword": "搜尋",
-    "Generate Info": "生成資訊",
-    "File Name": "檔案名",
-    "Collect": "收藏",
-    "Refresh page": "刷新頁面",
-    "Date to": "日期至",
-    "Move to favorites": "移動到收藏夾",
-    "Renew Page": "刷新頁面",
-    "Number": "數量",
-    "set_index": "設定索引",
-    "load_switch": "載入開關",
-    "turn_page_switch": "翻頁開關",
-    "Checkbox": "核取方塊",
-    "Apply settings": "儲存設定",
-    "Saving images/grids": "儲存圖像／宮格圖",
-    "Always save all generated images": "始終儲存所有生成的圖像",
-    "File format for images": "圖像的檔案格式",
-    "Images filename pattern": "圖像檔案名格式",
-    "Add number to filename when saving": "儲存的時候在檔案名里加入數字",
-    "Always save all generated image grids": "始終儲存所有生成的宮格圖",
-    "File format for grids": "宮格圖的檔案格式",
-    "Add extended info (seed, prompt) to filename when saving grid": "儲存宮格圖時，將擴展資訊（隨機種子，提示詞）加入到檔案名",
-    "Do not save grids consisting of one picture": "只有一張圖片時不要儲存宮格圖",
-    "Prevent empty spots in grid (when set to autodetect)": "（啟用自動偵測時）防止宮格圖中出現空位",
-    "Grid row count; use -1 for autodetect and 0 for it to be same as batch size": "宮格圖行數； 使用 -1 進行自動檢測，使用 0 使其與每批數量相同",
-    "Save text information about generation parameters as chunks to png files": "將有關生成參數的文本資訊，作為塊儲存到PNG圖片檔案中",
-    "Create a text file next to every image with generation parameters.": "儲存圖像時，在每個圖像旁邊建立一個文本檔案儲存生成參數",
-    "Save a copy of image before doing face restoration.": "在進行面部修復之前儲存圖像副本",
-    "Save a copy of image before applying highres fix.": "在做高解析度修復之前儲存初始圖像副本",
-    "Save a copy of image before applying color correction to img2img results": "在對圖生圖結果套用顏色校正之前儲存圖像副本",
-    "Quality for saved jpeg images": "儲存的JPEG圖像的品質",
-    "If PNG image is larger than 4MB or any dimension is larger than 4000, downscale and save copy as JPG": "如果 PNG 圖像大於 4MB 或寬高大於 4000，則縮小並儲存副本為 JPG 圖片",
-    "Use original name for output filename during batch process in extras tab": "在更多頁籤中的批量處理過程中，使用原始名稱作為輸出檔案名",
-    "When using 'Save' button, only save a single selected image": "使用「儲存」按鈕時，只儲存一個選定的圖像",
-    "Do not add watermark to images": "不要給圖像加浮水印",
-    "Paths for saving": "儲存路徑",
-    "Output directory for images; if empty, defaults to three directories below": "圖像的輸出目錄； 如果為空，則預設為以下三個目錄",
-    "Output directory for txt2img images": "文生圖的輸出目錄",
-    "Output directory for img2img images": "圖生圖的輸出目錄",
-    "Output directory for images from extras tab": "更多頁籤的輸出目錄",
-    "Output directory for grids; if empty, defaults to two directories below": "宮格圖的輸出目錄； 如果為空，則預設為以下兩個目錄",
-    "Output directory for txt2img grids": "文生圖宮格的輸出目錄",
-    "Output directory for img2img grids": "圖生圖宮格的輸出目錄",
-    "Directory for saving images using the Save button": "使用「儲存」按鈕儲存圖像的目錄",
-    "Saving to a directory": "儲存到目錄",
-    "Save images to a subdirectory": "將圖像儲存到子目錄",
-    "Save grids to a subdirectory": "將宮格圖儲存到子目錄",
-    "When using \"Save\" button, save images to a subdirectory": "使用「儲存」按鈕時，將圖像儲存到子目錄",
-    "Directory name pattern": "目錄名稱格式",
-    "Max prompt words for [prompt_words] pattern": "[prompt_words] 格式的最大提示詞數量",
-    "Upscaling": "放大",
-    "Tile size for ESRGAN upscalers. 0 = no tiling.": "ESRGAN 的圖塊尺寸。0 = 不分塊",
-    "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.": "ESRGAN 的圖塊重疊畫素。低值 = 可見接縫",
-    "Tile size for all SwinIR.": "適用所有 SwinIR 系演算法的圖塊尺寸",
-    "Tile overlap, in pixels for SwinIR. Low values = visible seam.": "SwinIR 的圖塊重疊畫素。低值 = 可見接縫",
-    "LDSR processing steps. Lower = faster": "LDSR 處理疊代步數。更低 = 更快",
-    "Upscaler for img2img": "圖生圖的放大演算法",
-    "Upscale latent space image when doing hires. fix": "做高解析度修復時，也放大潛空間圖像",
-    "Face restoration": "面部修復",
-    "CodeFormer weight parameter; 0 = maximum effect; 1 = minimum effect": "CodeFormer 權重參數； 0 = 最大效果； 1 = 最小效果",
-    "Move face restoration model from VRAM into RAM after processing": "面部修復處理完成後，將面部修復模型從顯存（VRAM）移至內存（RAM）",
-    "System": "系統",
-    "VRAM usage polls per second during generation. Set to 0 to disable.": "生成圖像時每秒輪詢顯存（VRAM）使用情況的次數。設定為 0 以禁用",
-    "Always print all generation info to standard output": "始終將所有生成資訊輸出到 standard output （一般為控制台）",
-    "Add a second progress bar to the console that shows progress for an entire job.": "向控制台加入第二個進度列，顯示整個作業的進度",
-    "Training": "訓練",
-    "Unload VAE and CLIP from VRAM when training": "訓練時從顯存（VRAM）中取消 VAE 和 CLIP 的載入",
-    "Move VAE and CLIP to RAM when training hypernetwork. Saves VRAM.": "訓練時將 VAE 和 CLIP 從顯存（VRAM）移放到內存（RAM），節省顯存（VRAM）",
-    "Move VAE and CLIP to RAM when training if possible. Saves VRAM.": "訓練時將 VAE 和 CLIP 從顯存（VRAM）移放到內存（RAM）如果可行的話，節省顯存（VRAM）",
-    "Filename word regex": "檔案名用詞的正則表達式",
-    "Filename join string": "檔案名連接用字串",
-    "Number of repeats for a single input image per epoch; used only for displaying epoch number": "每個 epoch 中單個輸入圖像的重複次數； 僅用於顯示 epoch 數",
-    "Save an csv containing the loss to log directory every N steps, 0 to disable": "每 N 步儲存一個包含 loss 的CSV表格到日誌目錄，0 表示禁用",
-    "Use cross attention optimizations while training": "訓練時開啟 cross attention 最佳化",
-    "Stable Diffusion": "Stable Diffusion",
-    "Checkpoints to cache in RAM": "快取在內存（RAM）中的模型權重存檔點",
-    "SD VAE": "模型的VAE",
-    "auto": "自動",
-    "Hypernetwork strength": "超網路（Hypernetwork）強度",
-    "Inpainting conditioning mask strength": "局部重繪時圖像調節的蒙版屏蔽強度",
-    "Apply color correction to img2img results to match original colors.": "對圖生圖結果套用顏色校正以匹配原始顏色",
-    "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising).": "在進行圖生圖的時候，確切地執行滑塊指定的疊代步數（正常情況下更弱的重繪幅度需要更少的疊代步數）",
-    "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.": "在 K 採樣器中啟用量化以獲得更清晰，更清晰的結果。這可能會改變現有的隨機種子。需要重新啟動才能套用",
-    "Emphasis: use (text) to make model pay more attention to text and [text] to make it pay less attention": "強調符：使用 (文字) 使模型更關注該文本，使用 [文字] 使其減少關注",
-    "Use old emphasis implementation. Can be useful to reproduce old seeds.": "使用舊的強調符實作。可用於復現舊隨機種子",
-    "Make K-diffusion samplers produce same images in a batch as when making a single image": "使 K-diffusion 採樣器批量生成與生成單個圖像時，產出相同的圖像",
-    "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens": "當使用超過 75 個 token 時，通過從 n 個 token 中的最後一個逗號填補來提高一致性",
-    "Filter NSFW content": "過濾成人內容",
-    "Stop At last layers of CLIP model": "在 CLIP 模型的最後哪一層停下",
-    "Interrogate Options": "反推提示詞選項",
-    "Interrogate: keep models in VRAM": "反推： 將模型儲存在顯存（VRAM）中",
-    "Interrogate: use artists from artists.csv": "反推： 使用 artists.csv 中的藝術家",
-    "Interrogate: include ranks of model tags matches in results (Has no effect on caption-based interrogators).": "反推： 在生成結果中包含與模型標記相匹配的等級（對基於生成自然語言描述的反推沒有影響）",
-    "Interrogate: num_beams for BLIP": "反推： BLIP 的 num_beams",
-    "Interrogate: minimum description length (excluding artists, etc..)": "反推： 最小描述長度（不包括藝術家，等…）",
-    "Interrogate: maximum description length": "反推： 最大描述長度",
-    "CLIP: maximum number of lines in text file (0 = No limit)": "CLIP： 文本檔案中的最大行數（0 = 無限制）",
-    "Interrogate: deepbooru score threshold": "反推： deepbooru 分數閾值",
-    "Interrogate: deepbooru sort alphabetically": "反推： deepbooru 按字母順序排序",
-    "use spaces for tags in deepbooru": "在 deepbooru 中為標記使用空格",
-    "escape (\\) brackets in deepbooru (so they are used as literal brackets and not for emphasis)": "在 deepbooru 中使用轉義 (\\) 括號（因此它們用作文字括號而不是強調符號）",
-    "User interface": "使用者介面",
-    "Show progressbar": "顯示進度列",
-    "Show image creation progress every N sampling steps. Set 0 to disable.": "每 N 個採樣疊代步數顯示圖像生成進度。設定 0 禁用",
-    "Show previews of all images generated in a batch as a grid": "以網格的形式，預覽批量生成的所有圖像",
-    "Show grid in results for web": "在網頁的結果中顯示宮格圖",
-    "Do not show any images in results for web": "不在網頁的結果中顯示任何圖像",
-    "Add model hash to generation information": "將模型的雜湊值加入到生成資訊",
-    "Add model name to generation information": "將模型名稱加入到生成資訊",
-    "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint.": "從文本讀取生成參數到使用者介面（從 PNG 圖片資訊或粘貼文本）時，不要更改選定的模型權重存檔點",
-    "Send seed when sending prompt or image to other interface": "將提示詞或者圖片發送到 >> 其他界面時，把隨機種子也傳送過去",
-    "Font for image grids that have text": "有文字的宮格圖使用的字體",
-    "Enable full page image viewer": "啟用整頁圖像檢視器",
-    "Show images zoomed in by default in full page image viewer": "在整頁圖像檢視器中，預設放大顯示圖像",
-    "Show generation progress in window title.": "在視窗標題中顯示生成進度",
-    "Quicksettings list": "快速設定列表",
-    "Localization (requires restart)": "本地化翻譯（需要儲存設定，並重啟Gradio）",
-    "Sampler parameters": "採樣器參數",
-    "Hide samplers in user interface (requires restart)": "在使用者介面中隱藏採樣器（需要重新啟動）",
-    "eta (noise multiplier) for DDIM": "DDIM 的 eta （噪聲乘數）",
-    "eta (noise multiplier) for ancestral samplers": "ancestral 採樣器的 eta （噪聲乘數）",
-    "img2img DDIM discretize": "圖生圖 DDIM 離散化",
-    "uniform": "均勻",
-    "quad": "二階",
-    "sigma churn": "sigma churn",
-    "sigma tmin": "最小(tmin) sigma",
-    "sigma noise": "sigma 噪聲",
-    "Eta noise seed delta": "Eta 噪聲種子偏移（ENSD）",
-    "Images Browser": "圖庫瀏覽器",
-    "Preload images at startup": "在啟動時預加載圖像",
-    "Number of columns on the page": "每頁列數",
-    "Number of rows on the page": "每頁行數",
-    "Number of pictures displayed on each page": "每頁顯示的圖像數量",
-    "Minimum number of pages per load": "每次載入的最小頁數",
-    "Number of grids in each row": "每行顯示多少格",
-    "Wildcards": "萬用字元",
-    "Use same seed for all images": "為所有圖像使用同一個隨機種子",
-    "Request browser notifications": "請求瀏覽器通知",
-    "Download localization template": "下載本地化模板",
-    "Reload custom script bodies (No ui updates, No restart)": "重新載入自訂指令碼主體（無使用者介面更新，無重啟）",
-    "Restart Gradio and Refresh components (Custom Scripts, ui.py, js and css only)": "重啟 Gradio 及刷新組件（僅限自訂指令碼，ui.py，JS 和 CSS）",
-    "Available": "可用",
-    "Install from URL": "從網址安裝",
-    "Apply and restart UI": "應用並重啟使用者介面",
-    "Check for updates": "檢查更新",
-    "Extension": "擴充",
-    "URL": "網址",
-    "Update": "更新",
-    "a1111-sd-webui-tagcomplete": "標記自動補全",
-    "unknown": "未知",
-    "deforum-for-automatic1111-webui": "Deforum",
-    "sd-dynamic-prompting": "動態提示詞",
-    "stable-diffusion-webui-aesthetic-gradients": "美術風格梯度",
-    "stable-diffusion-webui-aesthetic-image-scorer": "美術風格評等",
-    "stable-diffusion-webui-artists-to-study": "藝術家圖庫",
-    "stable-diffusion-webui-dataset-tag-editor": "資料集標記編輯器",
-    "stable-diffusion-webui-images-browser": "圖庫瀏覽器",
-    "stable-diffusion-webui-inspiration": "靈感",
-    "stable-diffusion-webui-wildcards": "萬用字元",
-    "Load from:": "載入自",
-    "Extension index URL": "擴充清單連結",
-    "URL for extension's git repository": "擴充的 git 倉庫連結",
-    "Local directory name": "本地路徑名",
-    "Prompt (press Ctrl+Enter or Alt+Enter to generate)": "提示詞（按 Ctrl+Enter 或 Alt+Enter 生成）",
-    "Negative prompt (press Ctrl+Enter or Alt+Enter to generate)": "反向提示詞（按 Ctrl+Enter 或 Alt+Enter 生成）",
-    "Add a random artist to the prompt.": "隨機加入一個藝術家到提示詞中",
-    "Read generation parameters from prompt or last generation if prompt is empty into user interface.": "從提示詞中讀取生成參數，如果提示詞為空，則讀取上一次的生成參數到使用者介面",
-    "Save style": "存儲為模板風格",
-    "Apply selected styles to current prompt": "將所選樣式套用於當前提示",
-    "Stop processing current image and continue processing.": "停止處理當前圖像，並繼續處理下一個",
-    "Stop processing images and return any results accumulated so far.": "停止處理圖像，並返回迄今為止累積的任何結果",
-    "Style to apply; styles have components for both positive and negative prompts and apply to both": "要套用的模版風格； 模版風格包含正向和反向提示詞，並套用於兩者",
-    "Do not do anything special": "什麼都不做",
-    "Which algorithm to use to produce the image": "使用哪種演算法生成圖像",
-    "Euler Ancestral - very creative, each can get a completely different picture depending on step count, setting steps to higher than 30-40 does not help": "Euler Ancestral - 非常有創意，可以根據疊代步數獲得完全不同的圖像，將疊代步數設定為高於 30-40 不會有正面作用",
-    "Denoising Diffusion Implicit Models - best at inpainting": "Denoising Diffusion Implicit models - 最擅長局部重繪",
-    "Produce an image that can be tiled.": "生成可用於平舖的圖像",
-    "Use a two step process to partially create an image at smaller resolution, upscale, and then improve details in it without changing composition": "使用兩步處理的時候，以較小的解析度生成初步圖像，接著放大圖像，然後在不更改構圖的情況下改進其中的細節",
-    "Determines how little respect the algorithm should have for image's content. At 0, nothing will change, and at 1 you'll get an unrelated image. With values below 1.0, processing will take less steps than the Sampling Steps slider specifies.": "決定演算法對圖像內容的影響程度。設定 0 時，什麼都不會改變，而在 1 時，你將獲得不相關的圖像。\n值低於 1.0 時，處理的疊代步數將少於「採樣疊代步數」滑塊指定的步數",
-    "How many batches of images to create": "建立多少批次的圖像",
-    "How many image to create in a single batch": "每批建立多少圖像",
-    "Classifier Free Guidance Scale - how strongly the image should conform to prompt - lower values produce more creative results": "Classifier Free Guidance Scale - 圖像應在多大程度上服從提示詞 - 較低的值會產生更有創意的結果",
-    "A value that determines the output of random number generator - if you create an image with same parameters and seed as another image, you'll get the same result": "一個固定隨機數生成器輸出的值 — 以相同參數和隨機種子生成的圖像會得到相同的結果",
-    "Set seed to -1, which will cause a new random number to be used every time": "將隨機種子設定為-1，則每次都會使用一個新的隨機數",
-    "Reuse seed from last generation, mostly useful if it was randomed": "重用上一次使用的隨機種子，如果想要固定結果就會很有用",
-    "Seed of a different picture to be mixed into the generation.": "將要參與生成的另一張圖的隨機種子",
-    "How strong of a variation to produce. At 0, there will be no effect. At 1, you will get the complete picture with variation seed (except for ancestral samplers, where you will just get something).": "想要產生多強烈的變化。設為 0 時，將沒有效果。設為 1 時，你將獲得完全產自差異隨機種子的圖像（ancestral 採樣器除外，你只是單純地生成了一些東西）",
-    "Make an attempt to produce a picture similar to what would have been produced with same seed at specified resolution": "嘗試生成與在指定解析度下使用相同隨機種子生成的圖像相似的圖片",
-    "This text is used to rotate the feature space of the imgs embs": "此文本用於旋轉圖集 embeddings 的特徵空間",
-    "Separate values for X axis using commas.": "使用逗號分隔 X 軸的值",
-    "Separate values for Y axis using commas.": "使用逗號分隔 Y 軸的值",
-    "Write image to a directory (default - log/images) and generation parameters into csv file.": "將圖像寫入目錄（預設 — log/images）並將生成參數寫入CSV表格檔案",
-    "Open images output directory": "打開圖像輸出目錄",
-    "How much to blur the mask before processing, in pixels.": "處理前要對蒙版進行多強的模糊，以畫素為單位",
-    "What to put inside the masked area before processing it with Stable Diffusion.": "在使用 Stable Diffusion 處理蒙版區域之前要在蒙版區域內放置什麼",
-    "fill it with colors of the image": "用圖像的顏色（高強度模糊）填充它",
-    "keep whatever was there originally": "保留原來的圖像，不進行預處理",
-    "fill it with latent space noise": "於潛空間填充噪聲",
-    "fill it with latent space zeroes": "於潛空間填零",
-    "Upscale masked region to target resolution, do inpainting, downscale back and paste into original image": "將蒙版區域（包括預留畫素長度的緩衝區域）放大到目標解析度，進行局部重繪。\n然後縮小並粘貼回原始圖像中",
-    "Resize image to target resolution. Unless height and width match, you will get incorrect aspect ratio.": "將圖像大小調整為目標解析度。除非高度和寬度匹配，否則你將獲得不正確的縱橫比",
-    "Resize the image so that entirety of target resolution is filled with the image. Crop parts that stick out.": "調整圖像大小，使整個目標解析度都被圖像填充。裁剪多出來的部分",
-    "Resize the image so that entirety of image is inside target resolution. Fill empty space with image's colors.": "調整圖像大小，使整個圖像在目標解析度內。用圖像的顏色填充空白區域",
-    "How many times to repeat processing an image and using it as input for the next iteration": "重複處理圖像並用作下次疊代輸入的次數",
-    "In loopback mode, on each loop the denoising strength is multiplied by this value. <1 means decreasing variety so your sequence will converge on a fixed picture. >1 means increasing variety so your sequence will become more and more chaotic.": "在回送模式下，在每個循環中，重繪幅度都會乘以該值。<1 表示減少多樣性，因此你的這一組圖將集中在固定的圖像上。>1 意味著增加多樣性，因此你的這一組圖將變得越來越混亂",
-    "For SD upscale, how much overlap in pixels should there be between tiles. Tiles overlap so that when they are merged back into one picture, there is no clearly visible seam.": "使用 SD 放大時，圖塊之間應該有多少畫素重疊。圖塊之間需要重疊才可以讓它們在合併回一張圖像時，沒有清晰可見的接縫",
-    "A directory on the same machine where the server is running.": "與伺服器主機上的目錄",
-    "Leave blank to save images to the default path.": "留空以將圖像儲存到預設路徑",
-    "Result = A * (1 - M) + B * M": "結果 = A * (1 - M) + B * M",
-    "Result = A + (B - C) * M": "結果 = A + (B - C) * M",
-    "1st and last digit must be 1. ex:'1, 2, 1'": "第一個和最後一個數字必須是 1。例：'1, 2, 1'",
-    "how fast should the training go. Low values will take longer to train, high values may fail to converge (not generate accurate results) and/or may break the embedding (This has happened if you see Loss: nan in the training info textbox. If this happens, you need to manually restore your embedding from an older not-broken backup).\n\nYou can set a single numeric value, or multiple learning rates using the syntax:\n\n   rate_1:max_steps_1, rate_2:max_steps_2, ...\n\nEG:   0.005:100, 1e-3:1000, 1e-5\n\nWill train with rate of 0.005 for first 100 steps, then 1e-3 until 1000 steps, then 1e-5 for all remaining steps.": "訓練應該多快。低值將需要更長的時間來訓練，高值可能無法收斂（無法產生準確的結果）以及／也許可能會破壞 embedding（如果你在訓練資訊文字方塊中看到 Loss: nan 就會發生這種情況。如果發生這種情況，你需要從較舊的未損壞的備份手動恢復 embedding）\n\n你可以使用以下語法設定單個數值或多個學習率：\n\n   率1:步限1, 率2:步限2, …\n\n如：   0.005:100, 1e-3:1000, 1e-5\n\n即前 100 步將以 0.005 的速率訓練，接著直到 1000 步為止以 1e-3 訓練，然後剩餘所有步以 1e-5 訓練",
-    "Path to directory with input images": "帶有輸入圖像的路徑",
-    "Path to directory where to write outputs": "進行輸出的路徑",
-    "Input images directory": "輸入圖像目錄",
-    "Use following tags to define how filenames for images are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "使用以下標記定義如何選擇圖像的檔案名： [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]； 預設請留空",
-    "If this option is enabled, watermark will not be added to created images. Warning: if you do not add watermark, you may be behaving in an unethical manner.": "如果啟用此選項，浮水印將不會加入到生成出來的圖像中。警告：如果你不加入浮水印，你的行為可能是不符合道德操守的",
-    "Use following tags to define how subdirectories for images and grids are chosen: [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]; leave empty for default.": "使用以下標記定義如何選擇圖像和宮格圖的子目錄： [steps], [cfg], [prompt], [prompt_no_styles], [prompt_spaces], [width], [height], [styles], [sampler], [seed], [model_hash], [prompt_words], [date], [datetime], [datetime<Format>], [datetime<Format><Time Zone>], [job_timestamp]； 預設請留空",
-    "Restore low quality faces using GFPGAN neural network": "使用 GFPGAN 神經網路修復低品質面部",
-    "This regular expression will be used extract words from filename, and they will be joined using the option below into label text used for training. Leave empty to keep filename text as it is.": "此正則表達式將用於從檔案名中提取單詞，並將使用以下選項將它們接合到用於訓練的標記文本中。留空以保持檔案名文本不變",
-    "This string will be used to join split words into a single line if the option above is enabled.": "如果啟用了上述選項，則此處的字元會用於將拆分的單詞接合為同一行",
-    "Only applies to inpainting models. Determines how strongly to mask off the original image for inpainting and img2img. 1.0 means fully masked, which is the default behaviour. 0.0 means a fully unmasked conditioning. Lower values will help preserve the overall composition of the image, but will struggle with large changes.": "僅適用於局部重繪專用的模型（模型後綴為 inpainting.ckpt 的模型）。決定了蒙版在局部重繪以及圖生圖中屏蔽原圖內容的強度。 1.0 表示完全屏蔽原圖，這是預設行為。 0.0 表示完全不屏蔽讓原圖進行圖像調節。較低的值將有助於保持原圖的整體構圖，但很難遇到較大的變化",
-    "List of setting names, separated by commas, for settings that should go to the quick access bar at the top, rather than the usual setting tab. See modules/shared.py for setting names. Requires restarting to apply.": "設定項名稱列表，以逗號分隔，該設定會移動到頂部的快速存取列，而不是預設的設定頁籤。有關設定名稱，請參見 modules/shared.py。需要重新啟動才能套用",
-    "If this values is non-zero, it will be added to seed and used to initialize RNG for noises when using samplers with Eta. You can use this to produce even more variation of images, or you can use this to match images of other software if you know what you are doing.": "如果這個值不為零，它將被加入到隨機種子中，並在使用帶有 Eta 的採樣器時用於初始化隨機噪聲。你可以使用它來產生更多的圖像變化，或者你可以使用它來模仿其他軟體生成的圖像，如果你知道你在做什麼",
-    "Leave empty for auto": "留空時自動生成",
-    "Autocomplete options": "自動補全選項",
-    "Enable Autocomplete": "開啟Tag補全",
-    "Select which Real-ESRGAN models to show in the web UI. (Requires restart)": "選擇哪些 Real-ESRGAN 模型顯示在網頁使用者介面。（需要重新啟動）",
-    "Allowed categories for random artists selection when using the Roll button": "使用抽選藝術家按鈕時將會隨機的藝術家類別",
-    "Append commas": "附加逗號",
-    "latest": "最新",
-    "behind": "落後",
-    "Roll three": "抽三位出來",
-    "Generate forever": "無限生成",
-    "Cancel generate forever": "停止無限生成",
-    "How many times to improve the generated image iteratively; higher values take longer; very low values can produce bad results": "疊代改進生成的圖像多少次；更高的值需要更長的時間；非常低的值會產生不好的結果",
-    "Draw a mask over an image, and the script will regenerate the masked area with content according to prompt": "在圖像上畫一個蒙版，指令碼會根據提示重新生成蒙版區域的內容",
-    "Upscale image normally, split result into tiles, improve each tile using img2img, merge whole image back": "正常放大圖像，將結果分割成圖塊，用圖生圖改進每個圖塊，最後將整個圖像合併回來",
-    "Create a grid where images will have different parameters. Use inputs below to specify which parameters will be shared by columns and rows": "創建一個網格，圖像將有不同的參數。使用下面的輸入來指定哪些參數將由列和行共享",
-    "Run Python code. Advanced user only. Must run program with --allow-code for this to work": "執行 Python 程式碼。僅限老手使用。必須以 --allow-code 來開啟程式，才能使其執行",
-    "Separate a list of words with commas, and the first word will be used as a keyword: script will search for this word in the prompt, and replace it with others": "以逗號分割的單詞列表，第一個單詞將被用作關鍵詞：指令碼將在提示詞中搜尋這個單詞，並用其他單詞替換它",
-    "Separate a list of words with commas, and the script will make a variation of prompt with those words for their every possible order": "以逗號分割的單詞列表，指令碼會排列出這些單詞的所有排列方式，並加入提示詞各生成一次",
-    "Reconstruct prompt from existing image and put it into the prompt field.": "從現有的圖像中重構出提示詞，並將其放入提示詞的輸入文字方塊",
-    "Set the maximum number of words to be used in the [prompt_words] option; ATTENTION: If the words are too long, they may exceed the maximum length of the file path that the system can handle": "設定在[prompt_words]選項中要使用的最大字數；注意：如果字數太長，可能會超過系統可處理的檔案路徑的最大長度",
-    "Process an image, use it as an input, repeat.": "處理一張圖像，將其作為輸入，並重複",
-    "Insert selected styles into prompt fields": "在提示詞中插入選定的模版風格",
-    "Save current prompts as a style. If you add the token {prompt} to the text, the style use that as placeholder for your prompt when you use the style in the future.": "將當前的提示詞儲存為模版風格。如果你在文本中加入{prompt}標記，那麼將來你使用該模版風格時，你現有的提示詞會替換模版風格中的{prompt}",
-    "Loads weights from checkpoint before making images. You can either use hash or a part of filename (as seen in settings) for checkpoint name. Recommended to use with Y axis for less switching.": "在生成圖像之前從模型權重存檔點中載入權重。你可以使用哈希值或檔案名的一部分（如設定中所示）作為模型權重存檔點名稱。建議用在Y軸上以減少過程中模型的切換",
-    "Torch active: Peak amount of VRAM used by Torch during generation, excluding cached data.\nTorch reserved: Peak amount of VRAM allocated by Torch, including all active and cached data.\nSys VRAM: Peak amount of VRAM allocation across all applications / total GPU VRAM (peak utilization%).": "Torch active: 在生成過程中，Torch使用的顯存(VRAM)峰值，不包括快取的數據。\nTorch reserved: Torch 分配的顯存(VRAM)的峰值量，包括所有活動和快取數據。\nSys VRAM: 所有應用程式分配的顯存(VRAM)的峰值量 / GPU 的總顯存(VRAM)（峰值利用率%）",
-    "Uscale the image in latent space. Alternative is to produce the full image from latent representation, upscale that, and then move it back to latent space.": "放大潛空間中的圖像。而另一種方法是，從潛變量表達中直接解碼並生成完整的圖像，接著放大它，然後再將其編碼回潛空間",
-    "Start drawing": "開始繪製",
-    "Description": "描述",
-    "Action": "行動",
-    "Aesthetic Gradients": "美術風格梯度",
-    "aesthetic-gradients": "美術風格梯度",
-    "Dynamic Prompts": "動態提示詞",
-    "images-browser": "圖庫瀏覽器",
-    "Inspiration": "靈感",
-    "Deforum": "Deforum",
-    "Artists to study": "藝術家圖庫",
-    "Aesthetic Image Scorer": "美術風格評分",
-    "Dataset Tag Editor": "數據集標記編輯器",
-    "Face restoration model": "面部修復模型",
-    "Install": "安裝",
-    "Installing...": "安裝中…",
-    "Installed": "已安裝"
-}

modules/api/api.py

@@ -3,14 +3,18 @@ import io
 import time
 import uvicorn
 from threading import Lock
-from gradio.processing_utils import encode_pil_to_base64, decode_base64_to_file, decode_base64_to_image
+from io import BytesIO
+from gradio.processing_utils import decode_base64_to_file
 from fastapi import APIRouter, Depends, FastAPI, HTTPException
+from fastapi.security import HTTPBasic, HTTPBasicCredentials
+from secrets import compare_digest
+
 import modules.shared as shared
+from modules import sd_samplers, deepbooru
 from modules.api.models import *
 from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, process_images
-from modules.sd_samplers import all_samplers
 from modules.extras import run_extras, run_pnginfo
-from PIL import PngImagePlugin
+from PIL import PngImagePlugin,Image
 from modules.sd_models import checkpoints_list
 from modules.realesrgan_model import get_realesrgan_models
 from typing import List
@@ -22,8 +26,12 @@ def upscaler_to_index(name: str):
         raise HTTPException(status_code=400, detail=f"Invalid upscaler, needs to be on of these: {' , '.join([x.name for x in sd_upscalers])}")
 
 
-sampler_to_index = lambda name: next(filter(lambda row: name.lower() == row[1].name.lower(), enumerate(all_samplers)), None)
+def validate_sampler_name(name):
+    config = sd_samplers.all_samplers_map.get(name, None)
+    if config is None:
+        raise HTTPException(status_code=404, detail="Sampler not found")
 
+    return name
 
 def setUpscalers(req: dict):
     reqDict = vars(req)
@@ -33,6 +41,10 @@ def setUpscalers(req: dict):
     reqDict.pop('upscaler_2')
     return reqDict
 
+def decode_base64_to_image(encoding):
+    if encoding.startswith("data:image/"):
+        encoding = encoding.split(";")[1].split(",")[1]
+    return Image.open(BytesIO(base64.b64decode(encoding)))
 
 def encode_pil_to_base64(image):
     with io.BytesIO() as output_bytes:
@@ -54,43 +66,59 @@ def encode_pil_to_base64(image):
 
 class Api:
     def __init__(self, app: FastAPI, queue_lock: Lock):
+        if shared.cmd_opts.api_auth:
+            self.credenticals = dict()
+            for auth in shared.cmd_opts.api_auth.split(","):
+                user, password = auth.split(":")
+                self.credenticals[user] = password
+
         self.router = APIRouter()
         self.app = app
         self.queue_lock = queue_lock
-        self.app.add_api_route("/sdapi/v1/txt2img", self.text2imgapi, methods=["POST"], response_model=TextToImageResponse)
-        self.app.add_api_route("/sdapi/v1/img2img", self.img2imgapi, methods=["POST"], response_model=ImageToImageResponse)
-        self.app.add_api_route("/sdapi/v1/extra-single-image", self.extras_single_image_api, methods=["POST"], response_model=ExtrasSingleImageResponse)
-        self.app.add_api_route("/sdapi/v1/extra-batch-images", self.extras_batch_images_api, methods=["POST"], response_model=ExtrasBatchImagesResponse)
-        self.app.add_api_route("/sdapi/v1/png-info", self.pnginfoapi, methods=["POST"], response_model=PNGInfoResponse)
-        self.app.add_api_route("/sdapi/v1/progress", self.progressapi, methods=["GET"], response_model=ProgressResponse)
-        self.app.add_api_route("/sdapi/v1/interrogate", self.interrogateapi, methods=["POST"])
-        self.app.add_api_route("/sdapi/v1/interrupt", self.interruptapi, methods=["POST"])
-        self.app.add_api_route("/sdapi/v1/options", self.get_config, methods=["GET"], response_model=OptionsModel)
-        self.app.add_api_route("/sdapi/v1/options", self.set_config, methods=["POST"])
-        self.app.add_api_route("/sdapi/v1/cmd-flags", self.get_cmd_flags, methods=["GET"], response_model=FlagsModel)
-        self.app.add_api_route("/sdapi/v1/samplers", self.get_samplers, methods=["GET"], response_model=List[SamplerItem])
-        self.app.add_api_route("/sdapi/v1/upscalers", self.get_upscalers, methods=["GET"], response_model=List[UpscalerItem])
-        self.app.add_api_route("/sdapi/v1/sd-models", self.get_sd_models, methods=["GET"], response_model=List[SDModelItem])
-        self.app.add_api_route("/sdapi/v1/hypernetworks", self.get_hypernetworks, methods=["GET"], response_model=List[HypernetworkItem])
-        self.app.add_api_route("/sdapi/v1/face-restorers", self.get_face_restorers, methods=["GET"], response_model=List[FaceRestorerItem])
-        self.app.add_api_route("/sdapi/v1/realesrgan-models", self.get_realesrgan_models, methods=["GET"], response_model=List[RealesrganItem])
-        self.app.add_api_route("/sdapi/v1/prompt-styles", self.get_promp_styles, methods=["GET"], response_model=List[PromptStyleItem])
-        self.app.add_api_route("/sdapi/v1/artist-categories", self.get_artists_categories, methods=["GET"], response_model=List[str])
-        self.app.add_api_route("/sdapi/v1/artists", self.get_artists, methods=["GET"], response_model=List[ArtistItem])
+        self.add_api_route("/sdapi/v1/txt2img", self.text2imgapi, methods=["POST"], response_model=TextToImageResponse)
+        self.add_api_route("/sdapi/v1/img2img", self.img2imgapi, methods=["POST"], response_model=ImageToImageResponse)
+        self.add_api_route("/sdapi/v1/extra-single-image", self.extras_single_image_api, methods=["POST"], response_model=ExtrasSingleImageResponse)
+        self.add_api_route("/sdapi/v1/extra-batch-images", self.extras_batch_images_api, methods=["POST"], response_model=ExtrasBatchImagesResponse)
+        self.add_api_route("/sdapi/v1/png-info", self.pnginfoapi, methods=["POST"], response_model=PNGInfoResponse)
+        self.add_api_route("/sdapi/v1/progress", self.progressapi, methods=["GET"], response_model=ProgressResponse)
+        self.add_api_route("/sdapi/v1/interrogate", self.interrogateapi, methods=["POST"])
+        self.add_api_route("/sdapi/v1/interrupt", self.interruptapi, methods=["POST"])
+        self.add_api_route("/sdapi/v1/skip", self.skip, methods=["POST"])
+        self.add_api_route("/sdapi/v1/options", self.get_config, methods=["GET"], response_model=OptionsModel)
+        self.add_api_route("/sdapi/v1/options", self.set_config, methods=["POST"])
+        self.add_api_route("/sdapi/v1/cmd-flags", self.get_cmd_flags, methods=["GET"], response_model=FlagsModel)
+        self.add_api_route("/sdapi/v1/samplers", self.get_samplers, methods=["GET"], response_model=List[SamplerItem])
+        self.add_api_route("/sdapi/v1/upscalers", self.get_upscalers, methods=["GET"], response_model=List[UpscalerItem])
+        self.add_api_route("/sdapi/v1/sd-models", self.get_sd_models, methods=["GET"], response_model=List[SDModelItem])
+        self.add_api_route("/sdapi/v1/hypernetworks", self.get_hypernetworks, methods=["GET"], response_model=List[HypernetworkItem])
+        self.add_api_route("/sdapi/v1/face-restorers", self.get_face_restorers, methods=["GET"], response_model=List[FaceRestorerItem])
+        self.add_api_route("/sdapi/v1/realesrgan-models", self.get_realesrgan_models, methods=["GET"], response_model=List[RealesrganItem])
+        self.add_api_route("/sdapi/v1/prompt-styles", self.get_promp_styles, methods=["GET"], response_model=List[PromptStyleItem])
+        self.add_api_route("/sdapi/v1/artist-categories", self.get_artists_categories, methods=["GET"], response_model=List[str])
+        self.add_api_route("/sdapi/v1/artists", self.get_artists, methods=["GET"], response_model=List[ArtistItem])
+
+    def add_api_route(self, path: str, endpoint, **kwargs):
+        if shared.cmd_opts.api_auth:
+            return self.app.add_api_route(path, endpoint, dependencies=[Depends(self.auth)], **kwargs)
+        return self.app.add_api_route(path, endpoint, **kwargs)
+
+    def auth(self, credenticals: HTTPBasicCredentials = Depends(HTTPBasic())):
+        if credenticals.username in self.credenticals:
+            if compare_digest(credenticals.password, self.credenticals[credenticals.username]):
+                return True
+
+        raise HTTPException(status_code=401, detail="Incorrect username or password", headers={"WWW-Authenticate": "Basic"})
 
     def text2imgapi(self, txt2imgreq: StableDiffusionTxt2ImgProcessingAPI):
-        sampler_index = sampler_to_index(txt2imgreq.sampler_index)
-
-        if sampler_index is None:
-            raise HTTPException(status_code=404, detail="Sampler not found")
-
         populate = txt2imgreq.copy(update={ # Override __init__ params
             "sd_model": shared.sd_model,
-            "sampler_index": sampler_index[0],
+            "sampler_name": validate_sampler_name(txt2imgreq.sampler_name or txt2imgreq.sampler_index),
             "do_not_save_samples": True,
             "do_not_save_grid": True
             }
         )
+        if populate.sampler_name:
+            populate.sampler_index = None  # prevent a warning later on
         p = StableDiffusionProcessingTxt2Img(**vars(populate))
         # Override object param
 
@@ -106,12 +134,6 @@ class Api:
         return TextToImageResponse(images=b64images, parameters=vars(txt2imgreq), info=processed.js())
 
     def img2imgapi(self, img2imgreq: StableDiffusionImg2ImgProcessingAPI):
-        sampler_index = sampler_to_index(img2imgreq.sampler_index)
-
-        if sampler_index is None:
-            raise HTTPException(status_code=404, detail="Sampler not found")
-
-
         init_images = img2imgreq.init_images
         if init_images is None:
             raise HTTPException(status_code=404, detail="Init image not found")
@@ -120,23 +142,22 @@ class Api:
         if mask:
             mask = decode_base64_to_image(mask)
 
-
         populate = img2imgreq.copy(update={ # Override __init__ params
             "sd_model": shared.sd_model,
-            "sampler_index": sampler_index[0],
+            "sampler_name": validate_sampler_name(img2imgreq.sampler_name or img2imgreq.sampler_index),
             "do_not_save_samples": True,
             "do_not_save_grid": True,
             "mask": mask
             }
         )
-        p = StableDiffusionProcessingImg2Img(**vars(populate))
+        if populate.sampler_name:
+            populate.sampler_index = None  # prevent a warning later on
 
-        imgs = []
-        for img in init_images:
-            img = decode_base64_to_image(img)
-            imgs = [img] * p.batch_size
+        args = vars(populate)
+        args.pop('include_init_images', None)  # this is meant to be done by "exclude": True in model, but it's for a reason that I cannot determine.
+        p = StableDiffusionProcessingImg2Img(**args)
 
-        p.init_images = imgs
+        p.init_images = [decode_base64_to_image(x) for x in init_images]
 
         shared.state.begin()
 
@@ -147,7 +168,7 @@ class Api:
 
         b64images = list(map(encode_pil_to_base64, processed.images))
 
-        if (not img2imgreq.include_init_images):
+        if not img2imgreq.include_init_images:
             img2imgreq.init_images = None
             img2imgreq.mask = None
 
@@ -220,11 +241,17 @@ class Api:
         if image_b64 is None:
             raise HTTPException(status_code=404, detail="Image not found") 
 
-        img = self.__base64_to_image(image_b64)
+        img = decode_base64_to_image(image_b64)
+        img = img.convert('RGB')
 
         # Override object param
         with self.queue_lock:
-            processed = shared.interrogator.interrogate(img)
+            if interrogatereq.model == "clip":
+                processed = shared.interrogator.interrogate(img)
+            elif interrogatereq.model == "deepdanbooru":
+                processed = deepbooru.model.tag(img)
+            else:
+                raise HTTPException(status_code=404, detail="Model not found")
         
         return InterrogateResponse(caption=processed)
 
@@ -233,6 +260,9 @@ class Api:
 
         return {}
 
+    def skip(self):
+        shared.state.skip()
+
     def get_config(self):
         options = {}
         for key in shared.opts.data.keys():
@@ -244,14 +274,9 @@ class Api:
 
         return options
 
-    def set_config(self, req: OptionsModel):
-        # currently req has all options fields even if you send a dict like { "send_seed": false }, which means it will
-        # overwrite all options with default values.
-        raise RuntimeError('Setting options via API is not supported')
-
-        reqDict = vars(req)
-        for o in reqDict:
-            setattr(shared.opts, o, reqDict[o])
+    def set_config(self, req: Dict[str, Any]):
+        for k, v in req.items():
+            shared.opts.set(k, v)
 
         shared.opts.save(shared.config_filename)
         return
@@ -260,7 +285,7 @@ class Api:
         return vars(shared.cmd_opts)
 
     def get_samplers(self):
-        return [{"name":sampler[0], "aliases":sampler[2], "options":sampler[3]} for sampler in all_samplers]
+        return [{"name": sampler[0], "aliases":sampler[2], "options":sampler[3]} for sampler in sd_samplers.all_samplers]
 
     def get_upscalers(self):
         upscalers = []
@@ -287,7 +312,7 @@ class Api:
         styleList = []
         for k in shared.prompt_styles.styles:
             style = shared.prompt_styles.styles[k]
-            styleList.append({"name":style[0], "prompt": style[1], "negative_prompr": style[2]})
+            styleList.append({"name":style[0], "prompt": style[1], "negative_prompt": style[2]})
 
         return styleList
 

modules/api/models.py

@@ -170,14 +170,15 @@ class ProgressResponse(BaseModel):
 
 class InterrogateRequest(BaseModel):
     image: str = Field(default="", title="Image", description="Image to work on, must be a Base64 string containing the image's data.")
+    model: str = Field(default="clip", title="Model", description="The interrogate model used.")
 
 class InterrogateResponse(BaseModel):
     caption: str = Field(default=None, title="Caption", description="The generated caption for the image.")
 
 fields = {}
-for key, value in opts.data.items():
-    metadata = opts.data_labels.get(key)
-    optType = opts.typemap.get(type(value), type(value))
+for key, metadata in opts.data_labels.items():
+    value = opts.data.get(key)
+    optType = opts.typemap.get(type(metadata.default), type(value))
 
     if (metadata is not None):
         fields.update({key: (Optional[optType], Field(

modules/call_queue.py

@@ -0,0 +1,98 @@
+import html
+import sys
+import threading
+import traceback
+import time
+
+from modules import shared
+
+queue_lock = threading.Lock()
+
+
+def wrap_queued_call(func):
+    def f(*args, **kwargs):
+        with queue_lock:
+            res = func(*args, **kwargs)
+
+        return res
+
+    return f
+
+
+def wrap_gradio_gpu_call(func, extra_outputs=None):
+    def f(*args, **kwargs):
+
+        shared.state.begin()
+
+        with queue_lock:
+            res = func(*args, **kwargs)
+
+        shared.state.end()
+
+        return res
+
+    return wrap_gradio_call(f, extra_outputs=extra_outputs, add_stats=True)
+
+
+def wrap_gradio_call(func, extra_outputs=None, add_stats=False):
+    def f(*args, extra_outputs_array=extra_outputs, **kwargs):
+        run_memmon = shared.opts.memmon_poll_rate > 0 and not shared.mem_mon.disabled and add_stats
+        if run_memmon:
+            shared.mem_mon.monitor()
+        t = time.perf_counter()
+
+        try:
+            res = list(func(*args, **kwargs))
+        except Exception as e:
+            # When printing out our debug argument list, do not print out more than a MB of text
+            max_debug_str_len = 131072 # (1024*1024)/8
+
+            print("Error completing request", file=sys.stderr)
+            argStr = f"Arguments: {str(args)} {str(kwargs)}"
+            print(argStr[:max_debug_str_len], file=sys.stderr)
+            if len(argStr) > max_debug_str_len:
+                print(f"(Argument list truncated at {max_debug_str_len}/{len(argStr)} characters)", file=sys.stderr)
+
+            print(traceback.format_exc(), file=sys.stderr)
+
+            shared.state.job = ""
+            shared.state.job_count = 0
+
+            if extra_outputs_array is None:
+                extra_outputs_array = [None, '']
+
+            res = extra_outputs_array + [f"<div class='error'>{html.escape(type(e).__name__+': '+str(e))}</div>"]
+
+        shared.state.skipped = False
+        shared.state.interrupted = False
+        shared.state.job_count = 0
+
+        if not add_stats:
+            return tuple(res)
+
+        elapsed = time.perf_counter() - t
+        elapsed_m = int(elapsed // 60)
+        elapsed_s = elapsed % 60
+        elapsed_text = f"{elapsed_s:.2f}s"
+        if elapsed_m > 0:
+            elapsed_text = f"{elapsed_m}m "+elapsed_text
+
+        if run_memmon:
+            mem_stats = {k: -(v//-(1024*1024)) for k, v in shared.mem_mon.stop().items()}
+            active_peak = mem_stats['active_peak']
+            reserved_peak = mem_stats['reserved_peak']
+            sys_peak = mem_stats['system_peak']
+            sys_total = mem_stats['total']
+            sys_pct = round(sys_peak/max(sys_total, 1) * 100, 2)
+
+            vram_html = f"<p class='vram'>Torch active/reserved: {active_peak}/{reserved_peak} MiB, <wbr>Sys VRAM: {sys_peak}/{sys_total} MiB ({sys_pct}%)</p>"
+        else:
+            vram_html = ''
+
+        # last item is always HTML
+        res[-1] += f"<div class='performance'><p class='time'>Time taken: <wbr>{elapsed_text}</p>{vram_html}</div>"
+
+        return tuple(res)
+
+    return f
+

modules/codeformer_model.py

@@ -36,6 +36,7 @@ def setup_model(dirname):
         from basicsr.utils.download_util import load_file_from_url
         from basicsr.utils import imwrite, img2tensor, tensor2img
         from facelib.utils.face_restoration_helper import FaceRestoreHelper
+        from facelib.detection.retinaface import retinaface
         from modules.shared import cmd_opts
 
         net_class = CodeFormer
@@ -65,6 +66,8 @@ def setup_model(dirname):
                 net.load_state_dict(checkpoint)
                 net.eval()
 
+                if hasattr(retinaface, 'device'):
+                    retinaface.device = devices.device_codeformer
                 face_helper = FaceRestoreHelper(1, face_size=512, crop_ratio=(1, 1), det_model='retinaface_resnet50', save_ext='png', use_parse=True, device=devices.device_codeformer)
 
                 self.net = net

modules/deepbooru.py

@@ -1,173 +1,97 @@
-import os.path
-from concurrent.futures import ProcessPoolExecutor
-import multiprocessing
-import time
+import os
 import re
 
+import torch
+from PIL import Image
+import numpy as np
+
+from modules import modelloader, paths, deepbooru_model, devices, images, shared
+
 re_special = re.compile(r'([\\()])')
 
-def get_deepbooru_tags(pil_image):
-    """
-    This method is for running only one image at a time for simple use.  Used to the img2img interrogate.
-    """
-    from modules import shared  # prevents circular reference
-
-    try:
-        create_deepbooru_process(shared.opts.interrogate_deepbooru_score_threshold, create_deepbooru_opts())
-        return get_tags_from_process(pil_image)
-    finally:
-        release_process()
-
-
-OPT_INCLUDE_RANKS = "include_ranks"
-def create_deepbooru_opts():
-    from modules import shared
-
-    return {
-        "use_spaces": shared.opts.deepbooru_use_spaces,
-        "use_escape": shared.opts.deepbooru_escape,
-        "alpha_sort": shared.opts.deepbooru_sort_alpha,
-        OPT_INCLUDE_RANKS: shared.opts.interrogate_return_ranks,
-    }
-
-
-def deepbooru_process(queue, deepbooru_process_return, threshold, deepbooru_opts):
-    model, tags = get_deepbooru_tags_model()
-    while True: # while process is running, keep monitoring queue for new image
-        pil_image = queue.get()
-        if pil_image == "QUIT":
-            break
-        else:
-            deepbooru_process_return["value"] = get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts)
-
-
-def create_deepbooru_process(threshold, deepbooru_opts):
-    """
-    Creates deepbooru process.  A queue is created to send images into the process.  This enables multiple images
-    to be processed in a row without reloading the model or creating a new process.  To return the data, a shared
-    dictionary is created to hold the tags created.  To wait for tags to be returned, a value of -1 is assigned
-    to the dictionary and the method adding the image to the queue should wait for this value to be updated with
-    the tags.
-    """
-    from modules import shared  # prevents circular reference
-    context = multiprocessing.get_context("spawn")
-    shared.deepbooru_process_manager = context.Manager()
-    shared.deepbooru_process_queue = shared.deepbooru_process_manager.Queue()
-    shared.deepbooru_process_return = shared.deepbooru_process_manager.dict()
-    shared.deepbooru_process_return["value"] = -1
-    shared.deepbooru_process = context.Process(target=deepbooru_process, args=(shared.deepbooru_process_queue, shared.deepbooru_process_return, threshold, deepbooru_opts))
-    shared.deepbooru_process.start()
-
-
-def get_tags_from_process(image):
-    from modules import shared
-
-    shared.deepbooru_process_return["value"] = -1
-    shared.deepbooru_process_queue.put(image)
-    while shared.deepbooru_process_return["value"] == -1:
-        time.sleep(0.2)
-    caption = shared.deepbooru_process_return["value"]
-    shared.deepbooru_process_return["value"] = -1
-
-    return caption
-
-
-def release_process():
-    """
-    Stops the deepbooru process to return used memory
-    """
-    from modules import shared  # prevents circular reference
-    shared.deepbooru_process_queue.put("QUIT")
-    shared.deepbooru_process.join()
-    shared.deepbooru_process_queue = None
-    shared.deepbooru_process = None
-    shared.deepbooru_process_return = None
-    shared.deepbooru_process_manager = None
-
-def get_deepbooru_tags_model():
-    import deepdanbooru as dd
-    import tensorflow as tf
-    import numpy as np
-    this_folder = os.path.dirname(__file__)
-    model_path = os.path.abspath(os.path.join(this_folder, '..', 'models', 'deepbooru'))
-    if not os.path.exists(os.path.join(model_path, 'project.json')):
-        # there is no point importing these every time
-        import zipfile
-        from basicsr.utils.download_util import load_file_from_url
-        load_file_from_url(
-            r"https://github.com/KichangKim/DeepDanbooru/releases/download/v3-20211112-sgd-e28/deepdanbooru-v3-20211112-sgd-e28.zip",
-            model_path)
-        with zipfile.ZipFile(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"), "r") as zip_ref:
-            zip_ref.extractall(model_path)
-        os.remove(os.path.join(model_path, "deepdanbooru-v3-20211112-sgd-e28.zip"))
-
-    tags = dd.project.load_tags_from_project(model_path)
-    model = dd.project.load_model_from_project(
-        model_path, compile_model=False
-    )
-    return model, tags
-
-
-def get_deepbooru_tags_from_model(model, tags, pil_image, threshold, deepbooru_opts):
-    import deepdanbooru as dd
-    import tensorflow as tf
-    import numpy as np
-
-    alpha_sort = deepbooru_opts['alpha_sort']
-    use_spaces = deepbooru_opts['use_spaces']
-    use_escape = deepbooru_opts['use_escape']
-    include_ranks = deepbooru_opts['include_ranks']
-
-    width = model.input_shape[2]
-    height = model.input_shape[1]
-    image = np.array(pil_image)
-    image = tf.image.resize(
-        image,
-        size=(height, width),
-        method=tf.image.ResizeMethod.AREA,
-        preserve_aspect_ratio=True,
-    )
-    image = image.numpy()  # EagerTensor to np.array
-    image = dd.image.transform_and_pad_image(image, width, height)
-    image = image / 255.0
-    image_shape = image.shape
-    image = image.reshape((1, image_shape[0], image_shape[1], image_shape[2]))
-
-    y = model.predict(image)[0]
-
-    result_dict = {}
-
-    for i, tag in enumerate(tags):
-        result_dict[tag] = y[i]
-
-    unsorted_tags_in_theshold = []
-    result_tags_print = []
-    for tag in tags:
-        if result_dict[tag] >= threshold:
+
+class DeepDanbooru:
+    def __init__(self):
+        self.model = None
+
+    def load(self):
+        if self.model is not None:
+            return
+
+        files = modelloader.load_models(
+            model_path=os.path.join(paths.models_path, "torch_deepdanbooru"),
+            model_url='https://github.com/AUTOMATIC1111/TorchDeepDanbooru/releases/download/v1/model-resnet_custom_v3.pt',
+            ext_filter=[".pt"],
+            download_name='model-resnet_custom_v3.pt',
+        )
+
+        self.model = deepbooru_model.DeepDanbooruModel()
+        self.model.load_state_dict(torch.load(files[0], map_location="cpu"))
+
+        self.model.eval()
+        self.model.to(devices.cpu, devices.dtype)
+
+    def start(self):
+        self.load()
+        self.model.to(devices.device)
+
+    def stop(self):
+        if not shared.opts.interrogate_keep_models_in_memory:
+            self.model.to(devices.cpu)
+            devices.torch_gc()
+
+    def tag(self, pil_image):
+        self.start()
+        res = self.tag_multi(pil_image)
+        self.stop()
+
+        return res
+
+    def tag_multi(self, pil_image, force_disable_ranks=False):
+        threshold = shared.opts.interrogate_deepbooru_score_threshold
+        use_spaces = shared.opts.deepbooru_use_spaces
+        use_escape = shared.opts.deepbooru_escape
+        alpha_sort = shared.opts.deepbooru_sort_alpha
+        include_ranks = shared.opts.interrogate_return_ranks and not force_disable_ranks
+
+        pic = images.resize_image(2, pil_image.convert("RGB"), 512, 512)
+        a = np.expand_dims(np.array(pic, dtype=np.float32), 0) / 255
+
+        with torch.no_grad(), devices.autocast():
+            x = torch.from_numpy(a).to(devices.device)
+            y = self.model(x)[0].detach().cpu().numpy()
+
+        probability_dict = {}
+
+        for tag, probability in zip(self.model.tags, y):
+            if probability < threshold:
+                continue
+
             if tag.startswith("rating:"):
                 continue
-            unsorted_tags_in_theshold.append((result_dict[tag], tag))
-            result_tags_print.append(f'{result_dict[tag]} {tag}')
-
-    # sort tags
-    result_tags_out = []
-    sort_ndx = 0
-    if alpha_sort:
-        sort_ndx = 1
-
-    # sort by reverse by likelihood and normal for alpha, and format tag text as requested
-    unsorted_tags_in_theshold.sort(key=lambda y: y[sort_ndx], reverse=(not alpha_sort))
-    for weight, tag in unsorted_tags_in_theshold:
-        tag_outformat = tag
-        if use_spaces:
-            tag_outformat = tag_outformat.replace('_', ' ')
-        if use_escape:
-            tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
-        if include_ranks:
-            tag_outformat = f"({tag_outformat}:{weight:.3f})"
-
-        result_tags_out.append(tag_outformat)
-
-    print('\n'.join(sorted(result_tags_print, reverse=True)))
-
-    return ', '.join(result_tags_out)
+
+            probability_dict[tag] = probability
+
+        if alpha_sort:
+            tags = sorted(probability_dict)
+        else:
+            tags = [tag for tag, _ in sorted(probability_dict.items(), key=lambda x: -x[1])]
+
+        res = []
+
+        for tag in tags:
+            probability = probability_dict[tag]
+            tag_outformat = tag
+            if use_spaces:
+                tag_outformat = tag_outformat.replace('_', ' ')
+            if use_escape:
+                tag_outformat = re.sub(re_special, r'\\\1', tag_outformat)
+            if include_ranks:
+                tag_outformat = f"({tag_outformat}:{probability:.3f})"
+
+            res.append(tag_outformat)
+
+        return ", ".join(res)
+
+
+model = DeepDanbooru()

modules/deepbooru_model.py

@@ -0,0 +1,676 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+# see https://github.com/AUTOMATIC1111/TorchDeepDanbooru for more
+
+
+class DeepDanbooruModel(nn.Module):
+    def __init__(self):
+        super(DeepDanbooruModel, self).__init__()
+
+        self.tags = []
+
+        self.n_Conv_0 = nn.Conv2d(kernel_size=(7, 7), in_channels=3, out_channels=64, stride=(2, 2))
+        self.n_MaxPool_0 = nn.MaxPool2d(kernel_size=(3, 3), stride=(2, 2))
+        self.n_Conv_1 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+        self.n_Conv_2 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=64)
+        self.n_Conv_3 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+        self.n_Conv_4 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+        self.n_Conv_5 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=64)
+        self.n_Conv_6 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+        self.n_Conv_7 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+        self.n_Conv_8 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=64)
+        self.n_Conv_9 = nn.Conv2d(kernel_size=(3, 3), in_channels=64, out_channels=64)
+        self.n_Conv_10 = nn.Conv2d(kernel_size=(1, 1), in_channels=64, out_channels=256)
+        self.n_Conv_11 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=512, stride=(2, 2))
+        self.n_Conv_12 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=128)
+        self.n_Conv_13 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128, stride=(2, 2))
+        self.n_Conv_14 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_15 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_16 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_17 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_18 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_19 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_20 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_21 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_22 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_23 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_24 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_25 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_26 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_27 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_28 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_29 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_30 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_31 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_32 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_33 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=128)
+        self.n_Conv_34 = nn.Conv2d(kernel_size=(3, 3), in_channels=128, out_channels=128)
+        self.n_Conv_35 = nn.Conv2d(kernel_size=(1, 1), in_channels=128, out_channels=512)
+        self.n_Conv_36 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=1024, stride=(2, 2))
+        self.n_Conv_37 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=256)
+        self.n_Conv_38 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256, stride=(2, 2))
+        self.n_Conv_39 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_40 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_41 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_42 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_43 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_44 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_45 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_46 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_47 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_48 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_49 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_50 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_51 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_52 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_53 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_54 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_55 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_56 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_57 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_58 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_59 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_60 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_61 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_62 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_63 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_64 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_65 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_66 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_67 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_68 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_69 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_70 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_71 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_72 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_73 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_74 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_75 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_76 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_77 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_78 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_79 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_80 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_81 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_82 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_83 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_84 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_85 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_86 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_87 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_88 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_89 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_90 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_91 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_92 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_93 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_94 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_95 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_96 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_97 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_98 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256, stride=(2, 2))
+        self.n_Conv_99 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_100 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=1024, stride=(2, 2))
+        self.n_Conv_101 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_102 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_103 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_104 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_105 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_106 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_107 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_108 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_109 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_110 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_111 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_112 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_113 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_114 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_115 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_116 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_117 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_118 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_119 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_120 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_121 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_122 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_123 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_124 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_125 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_126 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_127 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_128 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_129 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_130 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_131 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_132 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_133 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_134 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_135 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_136 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_137 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_138 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_139 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_140 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_141 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_142 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_143 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_144 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_145 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_146 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_147 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_148 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_149 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_150 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_151 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_152 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_153 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_154 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_155 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=256)
+        self.n_Conv_156 = nn.Conv2d(kernel_size=(3, 3), in_channels=256, out_channels=256)
+        self.n_Conv_157 = nn.Conv2d(kernel_size=(1, 1), in_channels=256, out_channels=1024)
+        self.n_Conv_158 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=2048, stride=(2, 2))
+        self.n_Conv_159 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=512)
+        self.n_Conv_160 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512, stride=(2, 2))
+        self.n_Conv_161 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+        self.n_Conv_162 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=512)
+        self.n_Conv_163 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512)
+        self.n_Conv_164 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+        self.n_Conv_165 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=512)
+        self.n_Conv_166 = nn.Conv2d(kernel_size=(3, 3), in_channels=512, out_channels=512)
+        self.n_Conv_167 = nn.Conv2d(kernel_size=(1, 1), in_channels=512, out_channels=2048)
+        self.n_Conv_168 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=4096, stride=(2, 2))
+        self.n_Conv_169 = nn.Conv2d(kernel_size=(1, 1), in_channels=2048, out_channels=1024)
+        self.n_Conv_170 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024, stride=(2, 2))
+        self.n_Conv_171 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+        self.n_Conv_172 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=1024)
+        self.n_Conv_173 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024)
+        self.n_Conv_174 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+        self.n_Conv_175 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=1024)
+        self.n_Conv_176 = nn.Conv2d(kernel_size=(3, 3), in_channels=1024, out_channels=1024)
+        self.n_Conv_177 = nn.Conv2d(kernel_size=(1, 1), in_channels=1024, out_channels=4096)
+        self.n_Conv_178 = nn.Conv2d(kernel_size=(1, 1), in_channels=4096, out_channels=9176, bias=False)
+
+    def forward(self, *inputs):
+        t_358, = inputs
+        t_359 = t_358.permute(*[0, 3, 1, 2])
+        t_359_padded = F.pad(t_359, [2, 3, 2, 3], value=0)
+        t_360 = self.n_Conv_0(t_359_padded)
+        t_361 = F.relu(t_360)
+        t_361 = F.pad(t_361, [0, 1, 0, 1], value=float('-inf'))
+        t_362 = self.n_MaxPool_0(t_361)
+        t_363 = self.n_Conv_1(t_362)
+        t_364 = self.n_Conv_2(t_362)
+        t_365 = F.relu(t_364)
+        t_365_padded = F.pad(t_365, [1, 1, 1, 1], value=0)
+        t_366 = self.n_Conv_3(t_365_padded)
+        t_367 = F.relu(t_366)
+        t_368 = self.n_Conv_4(t_367)
+        t_369 = torch.add(t_368, t_363)
+        t_370 = F.relu(t_369)
+        t_371 = self.n_Conv_5(t_370)
+        t_372 = F.relu(t_371)
+        t_372_padded = F.pad(t_372, [1, 1, 1, 1], value=0)
+        t_373 = self.n_Conv_6(t_372_padded)
+        t_374 = F.relu(t_373)
+        t_375 = self.n_Conv_7(t_374)
+        t_376 = torch.add(t_375, t_370)
+        t_377 = F.relu(t_376)
+        t_378 = self.n_Conv_8(t_377)
+        t_379 = F.relu(t_378)
+        t_379_padded = F.pad(t_379, [1, 1, 1, 1], value=0)
+        t_380 = self.n_Conv_9(t_379_padded)
+        t_381 = F.relu(t_380)
+        t_382 = self.n_Conv_10(t_381)
+        t_383 = torch.add(t_382, t_377)
+        t_384 = F.relu(t_383)
+        t_385 = self.n_Conv_11(t_384)
+        t_386 = self.n_Conv_12(t_384)
+        t_387 = F.relu(t_386)
+        t_387_padded = F.pad(t_387, [0, 1, 0, 1], value=0)
+        t_388 = self.n_Conv_13(t_387_padded)
+        t_389 = F.relu(t_388)
+        t_390 = self.n_Conv_14(t_389)
+        t_391 = torch.add(t_390, t_385)
+        t_392 = F.relu(t_391)
+        t_393 = self.n_Conv_15(t_392)
+        t_394 = F.relu(t_393)
+        t_394_padded = F.pad(t_394, [1, 1, 1, 1], value=0)
+        t_395 = self.n_Conv_16(t_394_padded)
+        t_396 = F.relu(t_395)
+        t_397 = self.n_Conv_17(t_396)
+        t_398 = torch.add(t_397, t_392)
+        t_399 = F.relu(t_398)
+        t_400 = self.n_Conv_18(t_399)
+        t_401 = F.relu(t_400)
+        t_401_padded = F.pad(t_401, [1, 1, 1, 1], value=0)
+        t_402 = self.n_Conv_19(t_401_padded)
+        t_403 = F.relu(t_402)
+        t_404 = self.n_Conv_20(t_403)
+        t_405 = torch.add(t_404, t_399)
+        t_406 = F.relu(t_405)
+        t_407 = self.n_Conv_21(t_406)
+        t_408 = F.relu(t_407)
+        t_408_padded = F.pad(t_408, [1, 1, 1, 1], value=0)
+        t_409 = self.n_Conv_22(t_408_padded)
+        t_410 = F.relu(t_409)
+        t_411 = self.n_Conv_23(t_410)
+        t_412 = torch.add(t_411, t_406)
+        t_413 = F.relu(t_412)
+        t_414 = self.n_Conv_24(t_413)
+        t_415 = F.relu(t_414)
+        t_415_padded = F.pad(t_415, [1, 1, 1, 1], value=0)
+        t_416 = self.n_Conv_25(t_415_padded)
+        t_417 = F.relu(t_416)
+        t_418 = self.n_Conv_26(t_417)
+        t_419 = torch.add(t_418, t_413)
+        t_420 = F.relu(t_419)
+        t_421 = self.n_Conv_27(t_420)
+        t_422 = F.relu(t_421)
+        t_422_padded = F.pad(t_422, [1, 1, 1, 1], value=0)
+        t_423 = self.n_Conv_28(t_422_padded)
+        t_424 = F.relu(t_423)
+        t_425 = self.n_Conv_29(t_424)
+        t_426 = torch.add(t_425, t_420)
+        t_427 = F.relu(t_426)
+        t_428 = self.n_Conv_30(t_427)
+        t_429 = F.relu(t_428)
+        t_429_padded = F.pad(t_429, [1, 1, 1, 1], value=0)
+        t_430 = self.n_Conv_31(t_429_padded)
+        t_431 = F.relu(t_430)
+        t_432 = self.n_Conv_32(t_431)
+        t_433 = torch.add(t_432, t_427)
+        t_434 = F.relu(t_433)
+        t_435 = self.n_Conv_33(t_434)
+        t_436 = F.relu(t_435)
+        t_436_padded = F.pad(t_436, [1, 1, 1, 1], value=0)
+        t_437 = self.n_Conv_34(t_436_padded)
+        t_438 = F.relu(t_437)
+        t_439 = self.n_Conv_35(t_438)
+        t_440 = torch.add(t_439, t_434)
+        t_441 = F.relu(t_440)
+        t_442 = self.n_Conv_36(t_441)
+        t_443 = self.n_Conv_37(t_441)
+        t_444 = F.relu(t_443)
+        t_444_padded = F.pad(t_444, [0, 1, 0, 1], value=0)
+        t_445 = self.n_Conv_38(t_444_padded)
+        t_446 = F.relu(t_445)
+        t_447 = self.n_Conv_39(t_446)
+        t_448 = torch.add(t_447, t_442)
+        t_449 = F.relu(t_448)
+        t_450 = self.n_Conv_40(t_449)
+        t_451 = F.relu(t_450)
+        t_451_padded = F.pad(t_451, [1, 1, 1, 1], value=0)
+        t_452 = self.n_Conv_41(t_451_padded)
+        t_453 = F.relu(t_452)
+        t_454 = self.n_Conv_42(t_453)
+        t_455 = torch.add(t_454, t_449)
+        t_456 = F.relu(t_455)
+        t_457 = self.n_Conv_43(t_456)
+        t_458 = F.relu(t_457)
+        t_458_padded = F.pad(t_458, [1, 1, 1, 1], value=0)
+        t_459 = self.n_Conv_44(t_458_padded)
+        t_460 = F.relu(t_459)
+        t_461 = self.n_Conv_45(t_460)
+        t_462 = torch.add(t_461, t_456)
+        t_463 = F.relu(t_462)
+        t_464 = self.n_Conv_46(t_463)
+        t_465 = F.relu(t_464)
+        t_465_padded = F.pad(t_465, [1, 1, 1, 1], value=0)
+        t_466 = self.n_Conv_47(t_465_padded)
+        t_467 = F.relu(t_466)
+        t_468 = self.n_Conv_48(t_467)
+        t_469 = torch.add(t_468, t_463)
+        t_470 = F.relu(t_469)
+        t_471 = self.n_Conv_49(t_470)
+        t_472 = F.relu(t_471)
+        t_472_padded = F.pad(t_472, [1, 1, 1, 1], value=0)
+        t_473 = self.n_Conv_50(t_472_padded)
+        t_474 = F.relu(t_473)
+        t_475 = self.n_Conv_51(t_474)
+        t_476 = torch.add(t_475, t_470)
+        t_477 = F.relu(t_476)
+        t_478 = self.n_Conv_52(t_477)
+        t_479 = F.relu(t_478)
+        t_479_padded = F.pad(t_479, [1, 1, 1, 1], value=0)
+        t_480 = self.n_Conv_53(t_479_padded)
+        t_481 = F.relu(t_480)
+        t_482 = self.n_Conv_54(t_481)
+        t_483 = torch.add(t_482, t_477)
+        t_484 = F.relu(t_483)
+        t_485 = self.n_Conv_55(t_484)
+        t_486 = F.relu(t_485)
+        t_486_padded = F.pad(t_486, [1, 1, 1, 1], value=0)
+        t_487 = self.n_Conv_56(t_486_padded)
+        t_488 = F.relu(t_487)
+        t_489 = self.n_Conv_57(t_488)
+        t_490 = torch.add(t_489, t_484)
+        t_491 = F.relu(t_490)
+        t_492 = self.n_Conv_58(t_491)
+        t_493 = F.relu(t_492)
+        t_493_padded = F.pad(t_493, [1, 1, 1, 1], value=0)
+        t_494 = self.n_Conv_59(t_493_padded)
+        t_495 = F.relu(t_494)
+        t_496 = self.n_Conv_60(t_495)
+        t_497 = torch.add(t_496, t_491)
+        t_498 = F.relu(t_497)
+        t_499 = self.n_Conv_61(t_498)
+        t_500 = F.relu(t_499)
+        t_500_padded = F.pad(t_500, [1, 1, 1, 1], value=0)
+        t_501 = self.n_Conv_62(t_500_padded)
+        t_502 = F.relu(t_501)
+        t_503 = self.n_Conv_63(t_502)
+        t_504 = torch.add(t_503, t_498)
+        t_505 = F.relu(t_504)
+        t_506 = self.n_Conv_64(t_505)
+        t_507 = F.relu(t_506)
+        t_507_padded = F.pad(t_507, [1, 1, 1, 1], value=0)
+        t_508 = self.n_Conv_65(t_507_padded)
+        t_509 = F.relu(t_508)
+        t_510 = self.n_Conv_66(t_509)
+        t_511 = torch.add(t_510, t_505)
+        t_512 = F.relu(t_511)
+        t_513 = self.n_Conv_67(t_512)
+        t_514 = F.relu(t_513)
+        t_514_padded = F.pad(t_514, [1, 1, 1, 1], value=0)
+        t_515 = self.n_Conv_68(t_514_padded)
+        t_516 = F.relu(t_515)
+        t_517 = self.n_Conv_69(t_516)
+        t_518 = torch.add(t_517, t_512)
+        t_519 = F.relu(t_518)
+        t_520 = self.n_Conv_70(t_519)
+        t_521 = F.relu(t_520)
+        t_521_padded = F.pad(t_521, [1, 1, 1, 1], value=0)
+        t_522 = self.n_Conv_71(t_521_padded)
+        t_523 = F.relu(t_522)
+        t_524 = self.n_Conv_72(t_523)
+        t_525 = torch.add(t_524, t_519)
+        t_526 = F.relu(t_525)
+        t_527 = self.n_Conv_73(t_526)
+        t_528 = F.relu(t_527)
+        t_528_padded = F.pad(t_528, [1, 1, 1, 1], value=0)
+        t_529 = self.n_Conv_74(t_528_padded)
+        t_530 = F.relu(t_529)
+        t_531 = self.n_Conv_75(t_530)
+        t_532 = torch.add(t_531, t_526)
+        t_533 = F.relu(t_532)
+        t_534 = self.n_Conv_76(t_533)
+        t_535 = F.relu(t_534)
+        t_535_padded = F.pad(t_535, [1, 1, 1, 1], value=0)
+        t_536 = self.n_Conv_77(t_535_padded)
+        t_537 = F.relu(t_536)
+        t_538 = self.n_Conv_78(t_537)
+        t_539 = torch.add(t_538, t_533)
+        t_540 = F.relu(t_539)
+        t_541 = self.n_Conv_79(t_540)
+        t_542 = F.relu(t_541)
+        t_542_padded = F.pad(t_542, [1, 1, 1, 1], value=0)
+        t_543 = self.n_Conv_80(t_542_padded)
+        t_544 = F.relu(t_543)
+        t_545 = self.n_Conv_81(t_544)
+        t_546 = torch.add(t_545, t_540)
+        t_547 = F.relu(t_546)
+        t_548 = self.n_Conv_82(t_547)
+        t_549 = F.relu(t_548)
+        t_549_padded = F.pad(t_549, [1, 1, 1, 1], value=0)
+        t_550 = self.n_Conv_83(t_549_padded)
+        t_551 = F.relu(t_550)
+        t_552 = self.n_Conv_84(t_551)
+        t_553 = torch.add(t_552, t_547)
+        t_554 = F.relu(t_553)
+        t_555 = self.n_Conv_85(t_554)
+        t_556 = F.relu(t_555)
+        t_556_padded = F.pad(t_556, [1, 1, 1, 1], value=0)
+        t_557 = self.n_Conv_86(t_556_padded)
+        t_558 = F.relu(t_557)
+        t_559 = self.n_Conv_87(t_558)
+        t_560 = torch.add(t_559, t_554)
+        t_561 = F.relu(t_560)
+        t_562 = self.n_Conv_88(t_561)
+        t_563 = F.relu(t_562)
+        t_563_padded = F.pad(t_563, [1, 1, 1, 1], value=0)
+        t_564 = self.n_Conv_89(t_563_padded)
+        t_565 = F.relu(t_564)
+        t_566 = self.n_Conv_90(t_565)
+        t_567 = torch.add(t_566, t_561)
+        t_568 = F.relu(t_567)
+        t_569 = self.n_Conv_91(t_568)
+        t_570 = F.relu(t_569)
+        t_570_padded = F.pad(t_570, [1, 1, 1, 1], value=0)
+        t_571 = self.n_Conv_92(t_570_padded)
+        t_572 = F.relu(t_571)
+        t_573 = self.n_Conv_93(t_572)
+        t_574 = torch.add(t_573, t_568)
+        t_575 = F.relu(t_574)
+        t_576 = self.n_Conv_94(t_575)
+        t_577 = F.relu(t_576)
+        t_577_padded = F.pad(t_577, [1, 1, 1, 1], value=0)
+        t_578 = self.n_Conv_95(t_577_padded)
+        t_579 = F.relu(t_578)
+        t_580 = self.n_Conv_96(t_579)
+        t_581 = torch.add(t_580, t_575)
+        t_582 = F.relu(t_581)
+        t_583 = self.n_Conv_97(t_582)
+        t_584 = F.relu(t_583)
+        t_584_padded = F.pad(t_584, [0, 1, 0, 1], value=0)
+        t_585 = self.n_Conv_98(t_584_padded)
+        t_586 = F.relu(t_585)
+        t_587 = self.n_Conv_99(t_586)
+        t_588 = self.n_Conv_100(t_582)
+        t_589 = torch.add(t_587, t_588)
+        t_590 = F.relu(t_589)
+        t_591 = self.n_Conv_101(t_590)
+        t_592 = F.relu(t_591)
+        t_592_padded = F.pad(t_592, [1, 1, 1, 1], value=0)
+        t_593 = self.n_Conv_102(t_592_padded)
+        t_594 = F.relu(t_593)
+        t_595 = self.n_Conv_103(t_594)
+        t_596 = torch.add(t_595, t_590)
+        t_597 = F.relu(t_596)
+        t_598 = self.n_Conv_104(t_597)
+        t_599 = F.relu(t_598)
+        t_599_padded = F.pad(t_599, [1, 1, 1, 1], value=0)
+        t_600 = self.n_Conv_105(t_599_padded)
+        t_601 = F.relu(t_600)
+        t_602 = self.n_Conv_106(t_601)
+        t_603 = torch.add(t_602, t_597)
+        t_604 = F.relu(t_603)
+        t_605 = self.n_Conv_107(t_604)
+        t_606 = F.relu(t_605)
+        t_606_padded = F.pad(t_606, [1, 1, 1, 1], value=0)
+        t_607 = self.n_Conv_108(t_606_padded)
+        t_608 = F.relu(t_607)
+        t_609 = self.n_Conv_109(t_608)
+        t_610 = torch.add(t_609, t_604)
+        t_611 = F.relu(t_610)
+        t_612 = self.n_Conv_110(t_611)
+        t_613 = F.relu(t_612)
+        t_613_padded = F.pad(t_613, [1, 1, 1, 1], value=0)
+        t_614 = self.n_Conv_111(t_613_padded)
+        t_615 = F.relu(t_614)
+        t_616 = self.n_Conv_112(t_615)
+        t_617 = torch.add(t_616, t_611)
+        t_618 = F.relu(t_617)
+        t_619 = self.n_Conv_113(t_618)
+        t_620 = F.relu(t_619)
+        t_620_padded = F.pad(t_620, [1, 1, 1, 1], value=0)
+        t_621 = self.n_Conv_114(t_620_padded)
+        t_622 = F.relu(t_621)
+        t_623 = self.n_Conv_115(t_622)
+        t_624 = torch.add(t_623, t_618)
+        t_625 = F.relu(t_624)
+        t_626 = self.n_Conv_116(t_625)
+        t_627 = F.relu(t_626)
+        t_627_padded = F.pad(t_627, [1, 1, 1, 1], value=0)
+        t_628 = self.n_Conv_117(t_627_padded)
+        t_629 = F.relu(t_628)
+        t_630 = self.n_Conv_118(t_629)
+        t_631 = torch.add(t_630, t_625)
+        t_632 = F.relu(t_631)
+        t_633 = self.n_Conv_119(t_632)
+        t_634 = F.relu(t_633)
+        t_634_padded = F.pad(t_634, [1, 1, 1, 1], value=0)
+        t_635 = self.n_Conv_120(t_634_padded)
+        t_636 = F.relu(t_635)
+        t_637 = self.n_Conv_121(t_636)
+        t_638 = torch.add(t_637, t_632)
+        t_639 = F.relu(t_638)
+        t_640 = self.n_Conv_122(t_639)
+        t_641 = F.relu(t_640)
+        t_641_padded = F.pad(t_641, [1, 1, 1, 1], value=0)
+        t_642 = self.n_Conv_123(t_641_padded)
+        t_643 = F.relu(t_642)
+        t_644 = self.n_Conv_124(t_643)
+        t_645 = torch.add(t_644, t_639)
+        t_646 = F.relu(t_645)
+        t_647 = self.n_Conv_125(t_646)
+        t_648 = F.relu(t_647)
+        t_648_padded = F.pad(t_648, [1, 1, 1, 1], value=0)
+        t_649 = self.n_Conv_126(t_648_padded)
+        t_650 = F.relu(t_649)
+        t_651 = self.n_Conv_127(t_650)
+        t_652 = torch.add(t_651, t_646)
+        t_653 = F.relu(t_652)
+        t_654 = self.n_Conv_128(t_653)
+        t_655 = F.relu(t_654)
+        t_655_padded = F.pad(t_655, [1, 1, 1, 1], value=0)
+        t_656 = self.n_Conv_129(t_655_padded)
+        t_657 = F.relu(t_656)
+        t_658 = self.n_Conv_130(t_657)
+        t_659 = torch.add(t_658, t_653)
+        t_660 = F.relu(t_659)
+        t_661 = self.n_Conv_131(t_660)
+        t_662 = F.relu(t_661)
+        t_662_padded = F.pad(t_662, [1, 1, 1, 1], value=0)
+        t_663 = self.n_Conv_132(t_662_padded)
+        t_664 = F.relu(t_663)
+        t_665 = self.n_Conv_133(t_664)
+        t_666 = torch.add(t_665, t_660)
+        t_667 = F.relu(t_666)
+        t_668 = self.n_Conv_134(t_667)
+        t_669 = F.relu(t_668)
+        t_669_padded = F.pad(t_669, [1, 1, 1, 1], value=0)
+        t_670 = self.n_Conv_135(t_669_padded)
+        t_671 = F.relu(t_670)
+        t_672 = self.n_Conv_136(t_671)
+        t_673 = torch.add(t_672, t_667)
+        t_674 = F.relu(t_673)
+        t_675 = self.n_Conv_137(t_674)
+        t_676 = F.relu(t_675)
+        t_676_padded = F.pad(t_676, [1, 1, 1, 1], value=0)
+        t_677 = self.n_Conv_138(t_676_padded)
+        t_678 = F.relu(t_677)
+        t_679 = self.n_Conv_139(t_678)
+        t_680 = torch.add(t_679, t_674)
+        t_681 = F.relu(t_680)
+        t_682 = self.n_Conv_140(t_681)
+        t_683 = F.relu(t_682)
+        t_683_padded = F.pad(t_683, [1, 1, 1, 1], value=0)
+        t_684 = self.n_Conv_141(t_683_padded)
+        t_685 = F.relu(t_684)
+        t_686 = self.n_Conv_142(t_685)
+        t_687 = torch.add(t_686, t_681)
+        t_688 = F.relu(t_687)
+        t_689 = self.n_Conv_143(t_688)
+        t_690 = F.relu(t_689)
+        t_690_padded = F.pad(t_690, [1, 1, 1, 1], value=0)
+        t_691 = self.n_Conv_144(t_690_padded)
+        t_692 = F.relu(t_691)
+        t_693 = self.n_Conv_145(t_692)
+        t_694 = torch.add(t_693, t_688)
+        t_695 = F.relu(t_694)
+        t_696 = self.n_Conv_146(t_695)
+        t_697 = F.relu(t_696)
+        t_697_padded = F.pad(t_697, [1, 1, 1, 1], value=0)
+        t_698 = self.n_Conv_147(t_697_padded)
+        t_699 = F.relu(t_698)
+        t_700 = self.n_Conv_148(t_699)
+        t_701 = torch.add(t_700, t_695)
+        t_702 = F.relu(t_701)
+        t_703 = self.n_Conv_149(t_702)
+        t_704 = F.relu(t_703)
+        t_704_padded = F.pad(t_704, [1, 1, 1, 1], value=0)
+        t_705 = self.n_Conv_150(t_704_padded)
+        t_706 = F.relu(t_705)
+        t_707 = self.n_Conv_151(t_706)
+        t_708 = torch.add(t_707, t_702)
+        t_709 = F.relu(t_708)
+        t_710 = self.n_Conv_152(t_709)
+        t_711 = F.relu(t_710)
+        t_711_padded = F.pad(t_711, [1, 1, 1, 1], value=0)
+        t_712 = self.n_Conv_153(t_711_padded)
+        t_713 = F.relu(t_712)
+        t_714 = self.n_Conv_154(t_713)
+        t_715 = torch.add(t_714, t_709)
+        t_716 = F.relu(t_715)
+        t_717 = self.n_Conv_155(t_716)
+        t_718 = F.relu(t_717)
+        t_718_padded = F.pad(t_718, [1, 1, 1, 1], value=0)
+        t_719 = self.n_Conv_156(t_718_padded)
+        t_720 = F.relu(t_719)
+        t_721 = self.n_Conv_157(t_720)
+        t_722 = torch.add(t_721, t_716)
+        t_723 = F.relu(t_722)
+        t_724 = self.n_Conv_158(t_723)
+        t_725 = self.n_Conv_159(t_723)
+        t_726 = F.relu(t_725)
+        t_726_padded = F.pad(t_726, [0, 1, 0, 1], value=0)
+        t_727 = self.n_Conv_160(t_726_padded)
+        t_728 = F.relu(t_727)
+        t_729 = self.n_Conv_161(t_728)
+        t_730 = torch.add(t_729, t_724)
+        t_731 = F.relu(t_730)
+        t_732 = self.n_Conv_162(t_731)
+        t_733 = F.relu(t_732)
+        t_733_padded = F.pad(t_733, [1, 1, 1, 1], value=0)
+        t_734 = self.n_Conv_163(t_733_padded)
+        t_735 = F.relu(t_734)
+        t_736 = self.n_Conv_164(t_735)
+        t_737 = torch.add(t_736, t_731)
+        t_738 = F.relu(t_737)
+        t_739 = self.n_Conv_165(t_738)
+        t_740 = F.relu(t_739)
+        t_740_padded = F.pad(t_740, [1, 1, 1, 1], value=0)
+        t_741 = self.n_Conv_166(t_740_padded)
+        t_742 = F.relu(t_741)
+        t_743 = self.n_Conv_167(t_742)
+        t_744 = torch.add(t_743, t_738)
+        t_745 = F.relu(t_744)
+        t_746 = self.n_Conv_168(t_745)
+        t_747 = self.n_Conv_169(t_745)
+        t_748 = F.relu(t_747)
+        t_748_padded = F.pad(t_748, [0, 1, 0, 1], value=0)
+        t_749 = self.n_Conv_170(t_748_padded)
+        t_750 = F.relu(t_749)
+        t_751 = self.n_Conv_171(t_750)
+        t_752 = torch.add(t_751, t_746)
+        t_753 = F.relu(t_752)
+        t_754 = self.n_Conv_172(t_753)
+        t_755 = F.relu(t_754)
+        t_755_padded = F.pad(t_755, [1, 1, 1, 1], value=0)
+        t_756 = self.n_Conv_173(t_755_padded)
+        t_757 = F.relu(t_756)
+        t_758 = self.n_Conv_174(t_757)
+        t_759 = torch.add(t_758, t_753)
+        t_760 = F.relu(t_759)
+        t_761 = self.n_Conv_175(t_760)
+        t_762 = F.relu(t_761)
+        t_762_padded = F.pad(t_762, [1, 1, 1, 1], value=0)
+        t_763 = self.n_Conv_176(t_762_padded)
+        t_764 = F.relu(t_763)
+        t_765 = self.n_Conv_177(t_764)
+        t_766 = torch.add(t_765, t_760)
+        t_767 = F.relu(t_766)
+        t_768 = self.n_Conv_178(t_767)
+        t_769 = F.avg_pool2d(t_768, kernel_size=t_768.shape[-2:])
+        t_770 = torch.squeeze(t_769, 3)
+        t_770 = torch.squeeze(t_770, 2)
+        t_771 = torch.sigmoid(t_770)
+        return t_771
+
+    def load_state_dict(self, state_dict, **kwargs):
+        self.tags = state_dict.get('tags', [])
+
+        super(DeepDanbooruModel, self).load_state_dict({k: v for k, v in state_dict.items() if k != 'tags'})
+

modules/devices.py

@@ -2,72 +2,95 @@ import sys, os, shlex
 import contextlib
 import torch
 from modules import errors
+from packaging import version
 
-# has_mps is only available in nightly pytorch (for now), `getattr` for compatibility
-has_mps = getattr(torch, 'has_mps', False)
 
-cpu = torch.device("cpu")
+# has_mps is only available in nightly pytorch (for now) and macOS 12.3+.
+# check `getattr` and try it for compatibility
+def has_mps() -> bool:
+    if not getattr(torch, 'has_mps', False):
+        return False
+    try:
+        torch.zeros(1).to(torch.device("mps"))
+        return True
+    except Exception:
+        return False
+
 
 def extract_device_id(args, name):
     for x in range(len(args)):
-        if name in args[x]: return args[x+1]
+        if name in args[x]:
+            return args[x + 1]
+
     return None
 
-def get_optimal_device():
-    if torch.cuda.is_available():
-        from modules import shared
 
-        device_id = shared.cmd_opts.device_id
+def get_cuda_device_string():
+    from modules import shared
+
+    if shared.cmd_opts.device_id is not None:
+        return f"cuda:{shared.cmd_opts.device_id}"
 
-        if device_id is not None:
-            cuda_device = f"cuda:{device_id}"
-            return torch.device(cuda_device)
-        else:
-            return torch.device("cuda")
+    return "cuda"
 
-    if has_mps:
+
+def get_optimal_device():
+    if torch.cuda.is_available():
+        return torch.device(get_cuda_device_string())
+
+    if has_mps():
         return torch.device("mps")
 
     return cpu
 
 
+def get_device_for(task):
+    from modules import shared
+
+    if task in shared.cmd_opts.use_cpu:
+        return cpu
+
+    return get_optimal_device()
+
+
 def torch_gc():
     if torch.cuda.is_available():
-        torch.cuda.empty_cache()
-        torch.cuda.ipc_collect()
+        with torch.cuda.device(get_cuda_device_string()):
+            torch.cuda.empty_cache()
+            torch.cuda.ipc_collect()
 
 
 def enable_tf32():
     if torch.cuda.is_available():
+
+        # enabling benchmark option seems to enable a range of cards to do fp16 when they otherwise can't
+        # see https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/4407
+        if any([torch.cuda.get_device_capability(devid) == (7, 5) for devid in range(0, torch.cuda.device_count())]):
+            torch.backends.cudnn.benchmark = True
+
         torch.backends.cuda.matmul.allow_tf32 = True
         torch.backends.cudnn.allow_tf32 = True
 
 
+
 errors.run(enable_tf32, "Enabling TF32")
 
-device = device_interrogate = device_gfpgan = device_swinir = device_esrgan = device_scunet = device_codeformer = None
+cpu = torch.device("cpu")
+device = device_interrogate = device_gfpgan = device_esrgan = device_codeformer = None
 dtype = torch.float16
 dtype_vae = torch.float16
 
-def randn(seed, shape):
-    # Pytorch currently doesn't handle setting randomness correctly when the metal backend is used.
-    if device.type == 'mps':
-        generator = torch.Generator(device=cpu)
-        generator.manual_seed(seed)
-        noise = torch.randn(shape, generator=generator, device=cpu).to(device)
-        return noise
 
+def randn(seed, shape):
     torch.manual_seed(seed)
+    if device.type == 'mps':
+        return torch.randn(shape, device=cpu).to(device)
     return torch.randn(shape, device=device)
 
 
 def randn_without_seed(shape):
-    # Pytorch currently doesn't handle setting randomness correctly when the metal backend is used.
     if device.type == 'mps':
-        generator = torch.Generator(device=cpu)
-        noise = torch.randn(shape, generator=generator, device=cpu).to(device)
-        return noise
-
+        return torch.randn(shape, device=cpu).to(device)
     return torch.randn(shape, device=device)
 
 
@@ -82,6 +105,27 @@ def autocast(disable=False):
 
     return torch.autocast("cuda")
 
+
 # MPS workaround for https://github.com/pytorch/pytorch/issues/79383
-def mps_contiguous(input_tensor, device): return input_tensor.contiguous() if device.type == 'mps' else input_tensor
-def mps_contiguous_to(input_tensor, device): return mps_contiguous(input_tensor, device).to(device)
+orig_tensor_to = torch.Tensor.to
+def tensor_to_fix(self, *args, **kwargs):
+    if self.device.type != 'mps' and \
+       ((len(args) > 0 and isinstance(args[0], torch.device) and args[0].type == 'mps') or \
+       (isinstance(kwargs.get('device'), torch.device) and kwargs['device'].type == 'mps')):
+        self = self.contiguous()
+    return orig_tensor_to(self, *args, **kwargs)
+
+
+# MPS workaround for https://github.com/pytorch/pytorch/issues/80800 
+orig_layer_norm = torch.nn.functional.layer_norm
+def layer_norm_fix(*args, **kwargs):
+    if len(args) > 0 and isinstance(args[0], torch.Tensor) and args[0].device.type == 'mps':
+        args = list(args)
+        args[0] = args[0].contiguous()
+    return orig_layer_norm(*args, **kwargs)
+
+
+# PyTorch 1.13 doesn't need these fixes but unfortunately is slower and has regressions that prevent training from working
+if has_mps() and version.parse(torch.__version__) < version.parse("1.13"):
+    torch.Tensor.to = tensor_to_fix
+    torch.nn.functional.layer_norm = layer_norm_fix

modules/esrgan_model.py

@@ -199,7 +199,7 @@ def upscale_without_tiling(model, img):
     img = img[:, :, ::-1]
     img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255
     img = torch.from_numpy(img).float()
-    img = devices.mps_contiguous_to(img.unsqueeze(0), devices.device_esrgan)
+    img = img.unsqueeze(0).to(devices.device_esrgan)
     with torch.no_grad():
         output = model(img)
     output = output.squeeze().float().cpu().clamp_(0, 1).numpy()

modules/extensions.py

@@ -6,9 +6,9 @@ import git
 
 from modules import paths, shared
 
-
 extensions = []
 extensions_dir = os.path.join(paths.script_path, "extensions")
+extensions_builtin_dir = os.path.join(paths.script_path, "extensions-builtin")
 
 
 def active():
@@ -16,12 +16,13 @@ def active():
 
 
 class Extension:
-    def __init__(self, name, path, enabled=True):
+    def __init__(self, name, path, enabled=True, is_builtin=False):
         self.name = name
         self.path = path
         self.enabled = enabled
         self.status = ''
         self.can_update = False
+        self.is_builtin = is_builtin
 
         repo = None
         try:
@@ -66,9 +67,12 @@ class Extension:
         self.can_update = False
         self.status = "latest"
 
-    def pull(self):
+    def fetch_and_reset_hard(self):
         repo = git.Repo(self.path)
-        repo.remotes.origin.pull()
+        # Fix: `error: Your local changes to the following files would be overwritten by merge`,
+        # because WSL2 Docker set 755 file permissions instead of 644, this results to the error.
+        repo.git.fetch('--all')
+        repo.git.reset('--hard', 'origin')
 
 
 def list_extensions():
@@ -77,10 +81,19 @@ def list_extensions():
     if not os.path.isdir(extensions_dir):
         return
 
-    for dirname in sorted(os.listdir(extensions_dir)):
-        path = os.path.join(extensions_dir, dirname)
-        if not os.path.isdir(path):
-            continue
+    paths = []
+    for dirname in [extensions_dir, extensions_builtin_dir]:
+        if not os.path.isdir(dirname):
+            return
+
+        for extension_dirname in sorted(os.listdir(dirname)):
+            path = os.path.join(dirname, extension_dirname)
+            if not os.path.isdir(path):
+                continue
 
-        extension = Extension(name=dirname, path=path, enabled=dirname not in shared.opts.disabled_extensions)
+            paths.append((extension_dirname, path, dirname == extensions_builtin_dir))
+
+    for dirname, path, is_builtin in paths:
+        extension = Extension(name=dirname, path=path, enabled=dirname not in shared.opts.disabled_extensions, is_builtin=is_builtin)
         extensions.append(extension)
+

modules/extras.py

@@ -1,6 +1,8 @@
 from __future__ import annotations
 import math
 import os
+import sys
+import traceback
 
 import numpy as np
 from PIL import Image
@@ -12,7 +14,7 @@ from typing import Callable, List, OrderedDict, Tuple
 from functools import partial
 from dataclasses import dataclass
 
-from modules import processing, shared, images, devices, sd_models
+from modules import processing, shared, images, devices, sd_models, sd_samplers
 from modules.shared import opts
 import modules.gfpgan_model
 from modules.ui import plaintext_to_html
@@ -20,7 +22,7 @@ import modules.codeformer_model
 import piexif
 import piexif.helper
 import gradio as gr
-
+import safetensors.torch
 
 class LruCache(OrderedDict):
     @dataclass(frozen=True)
@@ -60,7 +62,7 @@ def run_extras(extras_mode, resize_mode, image, image_folder, input_dir, output_
     # Also keep track of original file names
     imageNameArr = []
     outputs = []
-    
+
     if extras_mode == 1:
         #convert file to pillow image
         for img in image_folder:
@@ -213,25 +215,8 @@ def run_pnginfo(image):
     if image is None:
         return '', '', ''
 
-    items = image.info
-    geninfo = ''
-
-    if "exif" in image.info:
-        exif = piexif.load(image.info["exif"])
-        exif_comment = (exif or {}).get("Exif", {}).get(piexif.ExifIFD.UserComment, b'')
-        try:
-            exif_comment = piexif.helper.UserComment.load(exif_comment)
-        except ValueError:
-            exif_comment = exif_comment.decode('utf8', errors="ignore")
-
-        items['exif comment'] = exif_comment
-        geninfo = exif_comment
-
-        for field in ['jfif', 'jfif_version', 'jfif_unit', 'jfif_density', 'dpi', 'exif',
-                      'loop', 'background', 'timestamp', 'duration']:
-            items.pop(field, None)
-
-    geninfo = items.get('parameters', geninfo)
+    geninfo, items = images.read_info_from_image(image)
+    items = {**{'parameters': geninfo}, **items}
 
     info = ''
     for key, text in items.items():
@@ -249,7 +234,7 @@ def run_pnginfo(image):
     return '', geninfo, info
 
 
-def run_modelmerger(primary_model_name, secondary_model_name, teritary_model_name, interp_method, multiplier, save_as_half, custom_name):
+def run_modelmerger(primary_model_name, secondary_model_name, tertiary_model_name, interp_method, multiplier, save_as_half, custom_name, checkpoint_format):
     def weighted_sum(theta0, theta1, alpha):
         return ((1 - alpha) * theta0) + (alpha * theta1)
 
@@ -261,23 +246,8 @@ def run_modelmerger(primary_model_name, secondary_model_name, teritary_model_nam
 
     primary_model_info = sd_models.checkpoints_list[primary_model_name]
     secondary_model_info = sd_models.checkpoints_list[secondary_model_name]
-    teritary_model_info = sd_models.checkpoints_list.get(teritary_model_name, None)
-
-    print(f"Loading {primary_model_info.filename}...")
-    primary_model = torch.load(primary_model_info.filename, map_location='cpu')
-    theta_0 = sd_models.get_state_dict_from_checkpoint(primary_model)
-
-    print(f"Loading {secondary_model_info.filename}...")
-    secondary_model = torch.load(secondary_model_info.filename, map_location='cpu')
-    theta_1 = sd_models.get_state_dict_from_checkpoint(secondary_model)
-
-    if teritary_model_info is not None:
-        print(f"Loading {teritary_model_info.filename}...")
-        teritary_model = torch.load(teritary_model_info.filename, map_location='cpu')
-        theta_2 = sd_models.get_state_dict_from_checkpoint(teritary_model)
-    else:
-        teritary_model = None
-        theta_2 = None
+    tertiary_model_info = sd_models.checkpoints_list.get(tertiary_model_name, None)
+    result_is_inpainting_model = False
 
     theta_funcs = {
         "Weighted sum": (None, weighted_sum),
@@ -285,9 +255,16 @@ def run_modelmerger(primary_model_name, secondary_model_name, teritary_model_nam
     }
     theta_func1, theta_func2 = theta_funcs[interp_method]
 
-    print(f"Merging...")
+    if theta_func1 and not tertiary_model_info:
+        return ["Failed: Interpolation method requires a tertiary model."] + [gr.Dropdown.update(choices=sd_models.checkpoint_tiles()) for _ in range(4)]
+
+    print(f"Loading {secondary_model_info.filename}...")
+    theta_1 = sd_models.read_state_dict(secondary_model_info.filename, map_location='cpu')
 
     if theta_func1:
+        print(f"Loading {tertiary_model_info.filename}...")
+        theta_2 = sd_models.read_state_dict(tertiary_model_info.filename, map_location='cpu')
+
         for key in tqdm.tqdm(theta_1.keys()):
             if 'model' in key:
                 if key in theta_2:
@@ -295,12 +272,31 @@ def run_modelmerger(primary_model_name, secondary_model_name, teritary_model_nam
                     theta_1[key] = theta_func1(theta_1[key], t2)
                 else:
                     theta_1[key] = torch.zeros_like(theta_1[key])
-    del theta_2, teritary_model
+        del theta_2
+
+    print(f"Loading {primary_model_info.filename}...")
+    theta_0 = sd_models.read_state_dict(primary_model_info.filename, map_location='cpu')
+
+    print("Merging...")
 
     for key in tqdm.tqdm(theta_0.keys()):
         if 'model' in key and key in theta_1:
+            a = theta_0[key]
+            b = theta_1[key]
+
+            # this enables merging an inpainting model (A) with another one (B);
+            # where normal model would have 4 channels, for latenst space, inpainting model would
+            # have another 4 channels for unmasked picture's latent space, plus one channel for mask, for a total of 9
+            if a.shape != b.shape and a.shape[0:1] + a.shape[2:] == b.shape[0:1] + b.shape[2:]:
+                if a.shape[1] == 4 and b.shape[1] == 9:
+                    raise RuntimeError("When merging inpainting model with a normal one, A must be the inpainting model.")
 
-            theta_0[key] = theta_func2(theta_0[key], theta_1[key], multiplier)
+                assert a.shape[1] == 9 and b.shape[1] == 4, f"Bad dimensions for merged layer {key}: A={a.shape}, B={b.shape}"
+
+                theta_0[key][:, 0:4, :, :] = theta_func2(a[:, 0:4, :, :], b, multiplier)
+                result_is_inpainting_model = True
+            else:
+                theta_0[key] = theta_func2(a, b, multiplier)
 
             if save_as_half:
                 theta_0[key] = theta_0[key].half()
@@ -311,17 +307,31 @@ def run_modelmerger(primary_model_name, secondary_model_name, teritary_model_nam
             theta_0[key] = theta_1[key]
             if save_as_half:
                 theta_0[key] = theta_0[key].half()
+    del theta_1
 
     ckpt_dir = shared.cmd_opts.ckpt_dir or sd_models.model_path
 
-    filename = primary_model_info.model_name + '_' + str(round(1-multiplier, 2)) + '-' + secondary_model_info.model_name + '_' + str(round(multiplier, 2)) + '-' + interp_method.replace(" ", "_") + '-merged.ckpt'
-    filename = filename if custom_name == '' else (custom_name + '.ckpt')
+    filename = \
+        primary_model_info.model_name + '_' + str(round(1-multiplier, 2)) + '-' + \
+        secondary_model_info.model_name + '_' + str(round(multiplier, 2)) + '-' + \
+        interp_method.replace(" ", "_") + \
+        '-merged.' +  \
+        ("inpainting." if result_is_inpainting_model else "") + \
+        checkpoint_format
+
+    filename = filename if custom_name == '' else (custom_name + '.' + checkpoint_format)
+
     output_modelname = os.path.join(ckpt_dir, filename)
 
     print(f"Saving to {output_modelname}...")
-    torch.save(primary_model, output_modelname)
+
+    _, extension = os.path.splitext(output_modelname)
+    if extension.lower() == ".safetensors":
+        safetensors.torch.save_file(theta_0, output_modelname, metadata={"format": "pt"})
+    else:
+        torch.save(theta_0, output_modelname)
 
     sd_models.list_models()
 
-    print(f"Checkpoint saved.")
+    print("Checkpoint saved.")
     return ["Checkpoint saved to " + output_modelname] + [gr.Dropdown.update(choices=sd_models.checkpoint_tiles()) for _ in range(4)]

modules/generation_parameters_copypaste.py

@@ -2,6 +2,8 @@ import base64
 import io
 import os
 import re
+from pathlib import Path
+
 import gradio as gr
 from modules.shared import script_path
 from modules import shared
@@ -35,9 +37,8 @@ def quote(text):
 def image_from_url_text(filedata):
     if type(filedata) == dict and filedata["is_file"]:
         filename = filedata["name"]
-        tempdir = os.path.normpath(tempfile.gettempdir())
-        normfn = os.path.normpath(filename)
-        assert normfn.startswith(tempdir), 'trying to open image file not in temporary directory'
+        is_in_right_dir = any(Path(temp_dir).resolve() in Path(filename).resolve().parents for temp_dir in shared.demo.temp_dirs)
+        assert is_in_right_dir, 'trying to open image file outside of allowed directories'
 
         return Image.open(filename)
 
@@ -73,7 +74,10 @@ def integrate_settings_paste_fields(component_dict):
         'sd_hypernetwork': 'Hypernet',
         'sd_hypernetwork_strength': 'Hypernet strength',
         'CLIP_stop_at_last_layers': 'Clip skip',
+        'inpainting_mask_weight': 'Conditional mask weight',
         'sd_model_checkpoint': 'Model hash',
+        'eta_noise_seed_delta': 'ENSD',
+        'initial_noise_multiplier': 'Noise multiplier',
     }
     settings_paste_fields = [
         (component_dict[k], lambda d, k=k, v=v: ui.apply_setting(k, d.get(v, None)))
@@ -181,6 +185,10 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model
         else:
             res[k] = v
 
+    # Missing CLIP skip means it was set to 1 (the default)
+    if "Clip skip" not in res:
+        res["Clip skip"] = "1"
+
     return res
 
 

modules/gfpgan_model.py

@@ -36,7 +36,9 @@ def gfpgann():
     else:
         print("Unable to load gfpgan model!")
         return None
-    model = gfpgan_constructor(model_path=model_file, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None)
+    if hasattr(facexlib.detection.retinaface, 'device'):
+        facexlib.detection.retinaface.device = devices.device_gfpgan
+    model = gfpgan_constructor(model_path=model_file, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None, device=devices.device_gfpgan)
     loaded_gfpgan_model = model
 
     return model

modules/hypernetworks/hypernetwork.py

@@ -12,7 +12,7 @@ import torch
 import tqdm
 from einops import rearrange, repeat
 from ldm.util import default
-from modules import devices, processing, sd_models, shared
+from modules import devices, processing, sd_models, shared, sd_samplers
 from modules.textual_inversion import textual_inversion
 from modules.textual_inversion.learn_schedule import LearnRateScheduler
 from torch import einsum
@@ -38,7 +38,7 @@ class HypernetworkModule(torch.nn.Module):
     activation_dict.update({cls_name.lower(): cls_obj for cls_name, cls_obj in inspect.getmembers(torch.nn.modules.activation) if inspect.isclass(cls_obj) and cls_obj.__module__ == 'torch.nn.modules.activation'})
 
     def __init__(self, dim, state_dict=None, layer_structure=None, activation_func=None, weight_init='Normal',
-                 add_layer_norm=False, use_dropout=False, activate_output=False, last_layer_dropout=True):
+                 add_layer_norm=False, use_dropout=False, activate_output=False, last_layer_dropout=False):
         super().__init__()
 
         assert layer_structure is not None, "layer_structure must not be None"
@@ -154,16 +154,28 @@ class Hypernetwork:
                 HypernetworkModule(size, None, self.layer_structure, self.activation_func, self.weight_init,
                                    self.add_layer_norm, self.use_dropout, self.activate_output, last_layer_dropout=self.last_layer_dropout),
             )
+        self.eval_mode()
 
     def weights(self):
         res = []
+        for k, layers in self.layers.items():
+            for layer in layers:
+                res += layer.parameters()
+        return res
 
+    def train_mode(self):
         for k, layers in self.layers.items():
             for layer in layers:
                 layer.train()
-                res += layer.trainables()
+                for param in layer.parameters():
+                    param.requires_grad = True
 
-        return res
+    def eval_mode(self):
+        for k, layers in self.layers.items():
+            for layer in layers:
+                layer.eval()
+                for param in layer.parameters():
+                    param.requires_grad = False
 
     def save(self, filename):
         state_dict = {}
@@ -367,13 +379,13 @@ def report_statistics(loss_info:dict):
 
 
 
-def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log_directory, training_width, training_height, steps, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
+def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, steps, shuffle_tags, tag_drop_out, latent_sampling_method, create_image_every, save_hypernetwork_every, template_file, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
     # images allows training previews to have infotext. Importing it at the top causes a circular import problem.
     from modules import images
 
     save_hypernetwork_every = save_hypernetwork_every or 0
     create_image_every = create_image_every or 0
-    textual_inversion.validate_train_inputs(hypernetwork_name, learn_rate, batch_size, data_root, template_file, steps, save_hypernetwork_every, create_image_every, log_directory, name="hypernetwork")
+    textual_inversion.validate_train_inputs(hypernetwork_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_hypernetwork_every, create_image_every, log_directory, name="hypernetwork")
 
     path = shared.hypernetworks.get(hypernetwork_name, None)
     shared.loaded_hypernetwork = Hypernetwork()
@@ -403,32 +415,33 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log
     hypernetwork = shared.loaded_hypernetwork
     checkpoint = sd_models.select_checkpoint()
 
-    ititial_step = hypernetwork.step or 0
-    if ititial_step >= steps:
+    initial_step = hypernetwork.step or 0
+    if initial_step >= steps:
         shared.state.textinfo = f"Model has already been trained beyond specified max steps"
         return hypernetwork, filename
 
-    scheduler = LearnRateScheduler(learn_rate, steps, ititial_step)
-    
+    scheduler = LearnRateScheduler(learn_rate, steps, initial_step)
+
     # dataset loading may take a while, so input validations and early returns should be done before this
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
-    with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, device=devices.device, template_file=template_file, include_cond=True, batch_size=batch_size)
+
+    pin_memory = shared.opts.pin_memory
+
+    ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=hypernetwork_name, model=shared.sd_model, cond_model=shared.sd_model.cond_stage_model, device=devices.device, template_file=template_file, include_cond=True, batch_size=batch_size, gradient_step=gradient_step, shuffle_tags=shuffle_tags, tag_drop_out=tag_drop_out, latent_sampling_method=latent_sampling_method)
+    
+    latent_sampling_method = ds.latent_sampling_method
+
+    dl = modules.textual_inversion.dataset.PersonalizedDataLoader(ds, latent_sampling_method=latent_sampling_method, batch_size=ds.batch_size, pin_memory=pin_memory)
+
+    old_parallel_processing_allowed = shared.parallel_processing_allowed
 
     if unload:
+        shared.parallel_processing_allowed = False
         shared.sd_model.cond_stage_model.to(devices.cpu)
         shared.sd_model.first_stage_model.to(devices.cpu)
-
-    size = len(ds.indexes)
-    loss_dict = defaultdict(lambda : deque(maxlen = 1024))
-    losses = torch.zeros((size,))
-    previous_mean_losses = [0]
-    previous_mean_loss = 0
-    print("Mean loss of {} elements".format(size))
     
     weights = hypernetwork.weights()
-    for weight in weights:
-        weight.requires_grad = True
+    hypernetwork.train_mode()
 
     # Here we use optimizer from saved HN, or we can specify as UI option.
     if hypernetwork.optimizer_name in optimizer_dict:
@@ -446,139 +459,169 @@ def train_hypernetwork(hypernetwork_name, learn_rate, batch_size, data_root, log
             print("Cannot resume from saved optimizer!")
             print(e)
 
+    scaler = torch.cuda.amp.GradScaler()
+    
+    batch_size = ds.batch_size
+    gradient_step = ds.gradient_step
+    # n steps = batch_size * gradient_step * n image processed
+    steps_per_epoch = len(ds) // batch_size // gradient_step
+    max_steps_per_epoch = len(ds) // batch_size - (len(ds) // batch_size) % gradient_step
+    loss_step = 0
+    _loss_step = 0 #internal
+    # size = len(ds.indexes)
+    # loss_dict = defaultdict(lambda : deque(maxlen = 1024))
+    # losses = torch.zeros((size,))
+    # previous_mean_losses = [0]
+    # previous_mean_loss = 0
+    # print("Mean loss of {} elements".format(size))
+
     steps_without_grad = 0
 
     last_saved_file = "<none>"
     last_saved_image = "<none>"
     forced_filename = "<none>"
 
-    pbar = tqdm.tqdm(enumerate(ds), total=steps - ititial_step)
-    for i, entries in pbar:
-        hypernetwork.step = i + ititial_step
-        if len(loss_dict) > 0:
-            previous_mean_losses = [i[-1] for i in loss_dict.values()]
-            previous_mean_loss = mean(previous_mean_losses)
-            
-        scheduler.apply(optimizer, hypernetwork.step)
-        if scheduler.finished:
-            break
-
-        if shared.state.interrupted:
-            break
-
-        with torch.autocast("cuda"):
-            c = stack_conds([entry.cond for entry in entries]).to(devices.device)
-            # c = torch.vstack([entry.cond for entry in entries]).to(devices.device)
-            x = torch.stack([entry.latent for entry in entries]).to(devices.device)
-            loss = shared.sd_model(x, c)[0]
-            del x
-            del c
-
-            losses[hypernetwork.step % losses.shape[0]] = loss.item()
-            for entry in entries:
-                loss_dict[entry.filename].append(loss.item())
+    pbar = tqdm.tqdm(total=steps - initial_step)
+    try:
+        for i in range((steps-initial_step) * gradient_step):
+            if scheduler.finished:
+                break
+            if shared.state.interrupted:
+                break
+            for j, batch in enumerate(dl):
+                # works as a drop_last=True for gradient accumulation
+                if j == max_steps_per_epoch:
+                    break
+                scheduler.apply(optimizer, hypernetwork.step)
+                if scheduler.finished:
+                    break
+                if shared.state.interrupted:
+                    break
+
+                with devices.autocast():
+                    x = batch.latent_sample.to(devices.device, non_blocking=pin_memory)
+                    if tag_drop_out != 0 or shuffle_tags:
+                        shared.sd_model.cond_stage_model.to(devices.device)
+                        c = shared.sd_model.cond_stage_model(batch.cond_text).to(devices.device, non_blocking=pin_memory)
+                        shared.sd_model.cond_stage_model.to(devices.cpu)
+                    else:
+                        c = stack_conds(batch.cond).to(devices.device, non_blocking=pin_memory)
+                    loss = shared.sd_model(x, c)[0] / gradient_step
+                    del x
+                    del c
+
+                    _loss_step += loss.item()
+                scaler.scale(loss).backward()
+                # go back until we reach gradient accumulation steps
+                if (j + 1) % gradient_step != 0:
+                    continue
+                # print(f"grad:{weights[0].grad.detach().cpu().abs().mean().item():.7f}")
+                # scaler.unscale_(optimizer)
+                # print(f"grad:{weights[0].grad.detach().cpu().abs().mean().item():.15f}")
+                # torch.nn.utils.clip_grad_norm_(weights, max_norm=1.0)
+                # print(f"grad:{weights[0].grad.detach().cpu().abs().mean().item():.15f}")
+                scaler.step(optimizer)
+                scaler.update()
+                hypernetwork.step += 1
+                pbar.update()
+                optimizer.zero_grad(set_to_none=True)
+                loss_step = _loss_step
+                _loss_step = 0
+
+                steps_done = hypernetwork.step + 1
                 
-            optimizer.zero_grad()
-            weights[0].grad = None
-            loss.backward()
-
-            if weights[0].grad is None:
-                steps_without_grad += 1
-            else:
-                steps_without_grad = 0
-            assert steps_without_grad < 10, 'no gradient found for the trained weight after backward() for 10 steps in a row; this is a bug; training cannot continue'
-
-            optimizer.step()
-
-        steps_done = hypernetwork.step + 1
-
-        if torch.isnan(losses[hypernetwork.step % losses.shape[0]]): 
-            raise RuntimeError("Loss diverged.")
-        
-        if len(previous_mean_losses) > 1:
-            std = stdev(previous_mean_losses)
-        else:
-            std = 0
-        dataset_loss_info = f"dataset loss:{mean(previous_mean_losses):.3f}" + u"\u00B1" + f"({std / (len(previous_mean_losses) ** 0.5):.3f})"
-        pbar.set_description(dataset_loss_info)
-
-        if hypernetwork_dir is not None and steps_done % save_hypernetwork_every == 0:
-            # Before saving, change name to match current checkpoint.
-            hypernetwork_name_every = f'{hypernetwork_name}-{steps_done}'
-            last_saved_file = os.path.join(hypernetwork_dir, f'{hypernetwork_name_every}.pt')
-            hypernetwork.optimizer_name = optimizer_name
-            if shared.opts.save_optimizer_state:
-                hypernetwork.optimizer_state_dict = optimizer.state_dict()
-            save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, last_saved_file)
-            hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
-
-        textual_inversion.write_loss(log_directory, "hypernetwork_loss.csv", hypernetwork.step, len(ds), {
-            "loss": f"{previous_mean_loss:.7f}",
-            "learn_rate": scheduler.learn_rate
-        })
-
-        if images_dir is not None and steps_done % create_image_every == 0:
-            forced_filename = f'{hypernetwork_name}-{steps_done}'
-            last_saved_image = os.path.join(images_dir, forced_filename)
-
-            optimizer.zero_grad()
-            shared.sd_model.cond_stage_model.to(devices.device)
-            shared.sd_model.first_stage_model.to(devices.device)
-
-            p = processing.StableDiffusionProcessingTxt2Img(
-                sd_model=shared.sd_model,
-                do_not_save_grid=True,
-                do_not_save_samples=True,
-            )
-
-            if preview_from_txt2img:
-                p.prompt = preview_prompt
-                p.negative_prompt = preview_negative_prompt
-                p.steps = preview_steps
-                p.sampler_index = preview_sampler_index
-                p.cfg_scale = preview_cfg_scale
-                p.seed = preview_seed
-                p.width = preview_width
-                p.height = preview_height
-            else:
-                p.prompt = entries[0].cond_text
-                p.steps = 20
-
-            preview_text = p.prompt
+                epoch_num = hypernetwork.step // steps_per_epoch
+                epoch_step = hypernetwork.step % steps_per_epoch
+
+                pbar.set_description(f"[Epoch {epoch_num}: {epoch_step+1}/{steps_per_epoch}]loss: {loss_step:.7f}")
+                if hypernetwork_dir is not None and steps_done % save_hypernetwork_every == 0:
+                    # Before saving, change name to match current checkpoint.
+                    hypernetwork_name_every = f'{hypernetwork_name}-{steps_done}'
+                    last_saved_file = os.path.join(hypernetwork_dir, f'{hypernetwork_name_every}.pt')
+                    hypernetwork.optimizer_name = optimizer_name
+                    if shared.opts.save_optimizer_state:
+                        hypernetwork.optimizer_state_dict = optimizer.state_dict()
+                    save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, last_saved_file)
+                    hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
+
+                textual_inversion.write_loss(log_directory, "hypernetwork_loss.csv", hypernetwork.step, steps_per_epoch, {
+                    "loss": f"{loss_step:.7f}",
+                    "learn_rate": scheduler.learn_rate
+                })
+
+                if images_dir is not None and steps_done % create_image_every == 0:
+                    forced_filename = f'{hypernetwork_name}-{steps_done}'
+                    last_saved_image = os.path.join(images_dir, forced_filename)
+                    hypernetwork.eval_mode()
+                    shared.sd_model.cond_stage_model.to(devices.device)
+                    shared.sd_model.first_stage_model.to(devices.device)
+
+                    p = processing.StableDiffusionProcessingTxt2Img(
+                        sd_model=shared.sd_model,
+                        do_not_save_grid=True,
+                        do_not_save_samples=True,
+                    )
+
+                    if preview_from_txt2img:
+                        p.prompt = preview_prompt
+                        p.negative_prompt = preview_negative_prompt
+                        p.steps = preview_steps
+                        p.sampler_name = sd_samplers.samplers[preview_sampler_index].name
+                        p.cfg_scale = preview_cfg_scale
+                        p.seed = preview_seed
+                        p.width = preview_width
+                        p.height = preview_height
+                    else:
+                        p.prompt = batch.cond_text[0]
+                        p.steps = 20
+                        p.width = training_width
+                        p.height = training_height
 
-            processed = processing.process_images(p)
-            image = processed.images[0] if len(processed.images)>0 else None
+                    preview_text = p.prompt
 
-            if unload:
-                shared.sd_model.cond_stage_model.to(devices.cpu)
-                shared.sd_model.first_stage_model.to(devices.cpu)
+                    processed = processing.process_images(p)
+                    image = processed.images[0] if len(processed.images) > 0 else None
 
-            if image is not None:
-                shared.state.current_image = image
-                last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
-                last_saved_image += f", prompt: {preview_text}"
+                    if unload:
+                        shared.sd_model.cond_stage_model.to(devices.cpu)
+                        shared.sd_model.first_stage_model.to(devices.cpu)
+                    hypernetwork.train_mode()
+                    if image is not None:
+                        shared.state.current_image = image
+                        last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
+                        last_saved_image += f", prompt: {preview_text}"
 
-        shared.state.job_no = hypernetwork.step
+                shared.state.job_no = hypernetwork.step
 
-        shared.state.textinfo = f"""
+                shared.state.textinfo = f"""
 <p>
-Loss: {previous_mean_loss:.7f}<br/>
-Step: {hypernetwork.step}<br/>
-Last prompt: {html.escape(entries[0].cond_text)}<br/>
+Loss: {loss_step:.7f}<br/>
+Step: {steps_done}<br/>
+Last prompt: {html.escape(batch.cond_text[0])}<br/>
 Last saved hypernetwork: {html.escape(last_saved_file)}<br/>
 Last saved image: {html.escape(last_saved_image)}<br/>
 </p>
 """
-        
-    report_statistics(loss_dict)
+    except Exception:
+        print(traceback.format_exc(), file=sys.stderr)
+    finally:
+        pbar.leave = False
+        pbar.close()
+        hypernetwork.eval_mode()
+        #report_statistics(loss_dict)
 
     filename = os.path.join(shared.cmd_opts.hypernetwork_dir, f'{hypernetwork_name}.pt')
     hypernetwork.optimizer_name = optimizer_name
     if shared.opts.save_optimizer_state:
         hypernetwork.optimizer_state_dict = optimizer.state_dict()
     save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, filename)
+
     del optimizer
     hypernetwork.optimizer_state_dict = None  # dereference it after saving, to save memory.
+    shared.sd_model.cond_stage_model.to(devices.device)
+    shared.sd_model.first_stage_model.to(devices.device)
+    shared.parallel_processing_allowed = old_parallel_processing_allowed
+
     return hypernetwork, filename
 
 def save_hypernetwork(hypernetwork, checkpoint, hypernetwork_name, filename):

modules/images.py

@@ -15,6 +15,7 @@ import piexif.helper
 from PIL import Image, ImageFont, ImageDraw, PngImagePlugin
 from fonts.ttf import Roboto
 import string
+import json
 
 from modules import sd_samplers, shared, script_callbacks
 from modules.shared import opts, cmd_opts
@@ -303,8 +304,9 @@ class FilenameGenerator:
         'width': lambda self: self.image.width,
         'height': lambda self: self.image.height,
         'styles': lambda self: self.p and sanitize_filename_part(", ".join([style for style in self.p.styles if not style == "None"]) or "None", replace_spaces=False),
-        'sampler': lambda self: self.p and sanitize_filename_part(sd_samplers.samplers[self.p.sampler_index].name, replace_spaces=False),
+        'sampler': lambda self: self.p and sanitize_filename_part(self.p.sampler_name, replace_spaces=False),
         'model_hash': lambda self: getattr(self.p, "sd_model_hash", shared.sd_model.sd_model_hash),
+        'model_name': lambda self: sanitize_filename_part(shared.sd_model.sd_checkpoint_info.model_name, replace_spaces=False),
         'date': lambda self: datetime.datetime.now().strftime('%Y-%m-%d'),
         'datetime': lambda self, *args: self.datetime(*args),  # accepts formats: [datetime], [datetime<Format>], [datetime<Format><Time Zone>]
         'job_timestamp': lambda self: getattr(self.p, "job_timestamp", shared.state.job_timestamp),
@@ -524,6 +526,8 @@ def save_image(image, path, basename, seed=None, prompt=None, extension='png', i
     else:
         image.save(fullfn, quality=opts.jpeg_quality)
 
+    image.already_saved_as = fullfn
+
     target_side_length = 4000
     oversize = image.width > target_side_length or image.height > target_side_length
     if opts.export_for_4chan and (oversize or os.stat(fullfn).st_size > 4 * 1024 * 1024):
@@ -550,10 +554,45 @@ def save_image(image, path, basename, seed=None, prompt=None, extension='png', i
     return fullfn, txt_fullfn
 
 
+def read_info_from_image(image):
+    items = image.info or {}
+
+    geninfo = items.pop('parameters', None)
+
+    if "exif" in items:
+        exif = piexif.load(items["exif"])
+        exif_comment = (exif or {}).get("Exif", {}).get(piexif.ExifIFD.UserComment, b'')
+        try:
+            exif_comment = piexif.helper.UserComment.load(exif_comment)
+        except ValueError:
+            exif_comment = exif_comment.decode('utf8', errors="ignore")
+
+        items['exif comment'] = exif_comment
+        geninfo = exif_comment
+
+        for field in ['jfif', 'jfif_version', 'jfif_unit', 'jfif_density', 'dpi', 'exif',
+                      'loop', 'background', 'timestamp', 'duration']:
+            items.pop(field, None)
+
+    if items.get("Software", None) == "NovelAI":
+        try:
+            json_info = json.loads(items["Comment"])
+            sampler = sd_samplers.samplers_map.get(json_info["sampler"], "Euler a")
+
+            geninfo = f"""{items["Description"]}
+Negative prompt: {json_info["uc"]}
+Steps: {json_info["steps"]}, Sampler: {sampler}, CFG scale: {json_info["scale"]}, Seed: {json_info["seed"]}, Size: {image.width}x{image.height}, Clip skip: 2, ENSD: 31337"""
+        except Exception:
+            print(f"Error parsing NovelAI iamge generation parameters:", file=sys.stderr)
+            print(traceback.format_exc(), file=sys.stderr)
+
+    return geninfo, items
+
+
 def image_data(data):
     try:
         image = Image.open(io.BytesIO(data))
-        textinfo = image.text["parameters"]
+        textinfo, _ = read_info_from_image(image)
         return textinfo, None
     except Exception:
         pass

modules/img2img.py

@@ -4,9 +4,9 @@ import sys
 import traceback
 
 import numpy as np
-from PIL import Image, ImageOps, ImageChops
+from PIL import Image, ImageOps, ImageFilter, ImageEnhance, ImageChops
 
-from modules import devices
+from modules import devices, sd_samplers
 from modules.processing import Processed, StableDiffusionProcessingImg2Img, process_images
 from modules.shared import opts, state
 import modules.shared as shared
@@ -40,7 +40,7 @@ def process_batch(p, input_dir, output_dir, args):
 
         img = Image.open(image)
         # Use the EXIF orientation of photos taken by smartphones.
-        img = ImageOps.exif_transpose(img) 
+        img = ImageOps.exif_transpose(img)
         p.init_images = [img] * p.batch_size
 
         proc = modules.scripts.scripts_img2img.run(p, *args)
@@ -59,18 +59,31 @@ def process_batch(p, input_dir, output_dir, args):
                 processed_image.save(os.path.join(output_dir, filename))
 
 
-def img2img(mode: int, prompt: str, negative_prompt: str, prompt_style: str, prompt_style2: str, init_img, init_img_with_mask, init_img_inpaint, init_mask_inpaint, mask_mode, steps: int, sampler_index: int, mask_blur: int, inpainting_fill: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, *args):
+def img2img(mode: int, prompt: str, negative_prompt: str, prompt_style: str, prompt_style2: str, init_img, init_img_with_mask, init_img_with_mask_orig, init_img_inpaint, init_mask_inpaint, mask_mode, steps: int, sampler_index: int, mask_blur: int, mask_alpha: float, inpainting_fill: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, *args):
     is_inpaint = mode == 1
     is_batch = mode == 2
 
     if is_inpaint:
         # Drawn mask
         if mask_mode == 0:
-            image = init_img_with_mask['image']
-            mask = init_img_with_mask['mask']
-            alpha_mask = ImageOps.invert(image.split()[-1]).convert('L').point(lambda x: 255 if x > 0 else 0, mode='1')
-            mask = ImageChops.lighter(alpha_mask, mask.convert('L')).convert('L')
-            image = image.convert('RGB')
+            is_mask_sketch = isinstance(init_img_with_mask, dict)
+            is_mask_paint = not is_mask_sketch
+            if is_mask_sketch:
+                # Sketch: mask iff. not transparent
+                image, mask = init_img_with_mask["image"], init_img_with_mask["mask"]
+                alpha_mask = ImageOps.invert(image.split()[-1]).convert('L').point(lambda x: 255 if x > 0 else 0, mode='1')
+                mask = ImageChops.lighter(alpha_mask, mask.convert('L')).convert('L')
+            else:
+                # Color-sketch: mask iff. painted over
+                image = init_img_with_mask
+                orig = init_img_with_mask_orig or init_img_with_mask
+                pred = np.any(np.array(image) != np.array(orig), axis=-1)
+                mask = Image.fromarray(pred.astype(np.uint8) * 255, "L")
+                mask = ImageEnhance.Brightness(mask).enhance(1 - mask_alpha / 100)
+                blur = ImageFilter.GaussianBlur(mask_blur)
+                image = Image.composite(image.filter(blur), orig, mask.filter(blur))
+
+            image = image.convert("RGB")
         # Uploaded mask
         else:
             image = init_img_inpaint
@@ -82,7 +95,7 @@ def img2img(mode: int, prompt: str, negative_prompt: str, prompt_style: str, pro
 
     # Use the EXIF orientation of photos taken by smartphones.
     if image is not None:
-        image = ImageOps.exif_transpose(image) 
+        image = ImageOps.exif_transpose(image)
 
     assert 0. <= denoising_strength <= 1., 'can only work with strength in [0.0, 1.0]'
 
@@ -99,7 +112,7 @@ def img2img(mode: int, prompt: str, negative_prompt: str, prompt_style: str, pro
         seed_resize_from_h=seed_resize_from_h,
         seed_resize_from_w=seed_resize_from_w,
         seed_enable_extras=seed_enable_extras,
-        sampler_index=sampler_index,
+        sampler_name=sd_samplers.samplers_for_img2img[sampler_index].name,
         batch_size=batch_size,
         n_iter=n_iter,
         steps=steps,

modules/interrogate.py

@@ -1,4 +1,3 @@
-import contextlib
 import os
 import sys
 import traceback
@@ -11,10 +10,9 @@ from torchvision import transforms
 from torchvision.transforms.functional import InterpolationMode
 
 import modules.shared as shared
-from modules import devices, paths, lowvram
+from modules import devices, paths, lowvram, modelloader
 
 blip_image_eval_size = 384
-blip_model_url = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_caption_capfilt_large.pth'
 clip_model_name = 'ViT-L/14'
 
 Category = namedtuple("Category", ["name", "topn", "items"])
@@ -47,7 +45,14 @@ class InterrogateModels:
     def load_blip_model(self):
         import models.blip
 
-        blip_model = models.blip.blip_decoder(pretrained=blip_model_url, image_size=blip_image_eval_size, vit='base', med_config=os.path.join(paths.paths["BLIP"], "configs", "med_config.json"))
+        files = modelloader.load_models(
+            model_path=os.path.join(paths.models_path, "BLIP"),
+            model_url='https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_caption_capfilt_large.pth',
+            ext_filter=[".pth"],
+            download_name='model_base_caption_capfilt_large.pth',
+        )
+
+        blip_model = models.blip.blip_decoder(pretrained=files[0], image_size=blip_image_eval_size, vit='base', med_config=os.path.join(paths.paths["BLIP"], "configs", "med_config.json"))
         blip_model.eval()
 
         return blip_model
@@ -148,8 +153,7 @@ class InterrogateModels:
 
             clip_image = self.clip_preprocess(pil_image).unsqueeze(0).type(self.dtype).to(devices.device_interrogate)
 
-            precision_scope = torch.autocast if shared.cmd_opts.precision == "autocast" else contextlib.nullcontext
-            with torch.no_grad(), precision_scope("cuda"):
+            with torch.no_grad(), devices.autocast():
                 image_features = self.clip_model.encode_image(clip_image).type(self.dtype)
 
                 image_features /= image_features.norm(dim=-1, keepdim=True)

modules/lowvram.py

@@ -51,6 +51,10 @@ def setup_for_low_vram(sd_model, use_medvram):
         send_me_to_gpu(first_stage_model, None)
         return first_stage_model_decode(z)
 
+    # for SD1, cond_stage_model is CLIP and its NN is in the tranformer frield, but for SD2, it's open clip, and it's in model field
+    if hasattr(sd_model.cond_stage_model, 'model'):
+        sd_model.cond_stage_model.transformer = sd_model.cond_stage_model.model
+
     # remove three big modules, cond, first_stage, and unet from the model and then
     # send the model to GPU. Then put modules back. the modules will be in CPU.
     stored = sd_model.cond_stage_model.transformer, sd_model.first_stage_model, sd_model.model
@@ -65,6 +69,10 @@ def setup_for_low_vram(sd_model, use_medvram):
     sd_model.first_stage_model.decode = first_stage_model_decode_wrap
     parents[sd_model.cond_stage_model.transformer] = sd_model.cond_stage_model
 
+    if hasattr(sd_model.cond_stage_model, 'model'):
+        sd_model.cond_stage_model.model = sd_model.cond_stage_model.transformer
+        del sd_model.cond_stage_model.transformer
+
     if use_medvram:
         sd_model.model.register_forward_pre_hook(send_me_to_gpu)
     else:

modules/modelloader.py

@@ -82,6 +82,7 @@ def cleanup_models():
     src_path = models_path
     dest_path = os.path.join(models_path, "Stable-diffusion")
     move_files(src_path, dest_path, ".ckpt")
+    move_files(src_path, dest_path, ".safetensors")
     src_path = os.path.join(root_path, "ESRGAN")
     dest_path = os.path.join(models_path, "ESRGAN")
     move_files(src_path, dest_path)
@@ -123,10 +124,9 @@ def move_files(src_path: str, dest_path: str, ext_filter: str = None):
 
 
 def load_upscalers():
-    sd = shared.script_path
     # We can only do this 'magic' method to dynamically load upscalers if they are referenced,
     # so we'll try to import any _model.py files before looking in __subclasses__
-    modules_dir = os.path.join(sd, "modules")
+    modules_dir = os.path.join(shared.script_path, "modules")
     for file in os.listdir(modules_dir):
         if "_model.py" in file:
             model_name = file.replace("_model.py", "")
@@ -135,22 +135,13 @@ def load_upscalers():
                 importlib.import_module(full_model)
             except:
                 pass
+
     datas = []
-    c_o = vars(shared.cmd_opts)
+    commandline_options = vars(shared.cmd_opts)
     for cls in Upscaler.__subclasses__():
         name = cls.__name__
-        module_name = cls.__module__
-        module = importlib.import_module(module_name)
-        class_ = getattr(module, name)
         cmd_name = f"{name.lower().replace('upscaler', '')}_models_path"
-        opt_string = None
-        try:
-            if cmd_name in c_o:
-                opt_string = c_o[cmd_name]
-        except:
-            pass
-        scaler = class_(opt_string)
-        for child in scaler.scalers:
-            datas.append(child)
+        scaler = cls(commandline_options.get(cmd_name, None))
+        datas += scaler.scalers
 
     shared.sd_upscalers = datas

modules/ngrok.py

@@ -1,14 +1,23 @@
 from pyngrok import ngrok, conf, exception
 
-
 def connect(token, port, region):
+    account = None
     if token == None:
         token = 'None'
+    else:
+        if ':' in token:
+            # token = authtoken:username:password
+            account = token.split(':')[1] + ':' + token.split(':')[-1]
+            token = token.split(':')[0]
+
     config = conf.PyngrokConfig(
         auth_token=token, region=region
     )
     try:
-        public_url = ngrok.connect(port, pyngrok_config=config).public_url
+        if account == None:
+            public_url = ngrok.connect(port, pyngrok_config=config, bind_tls=True).public_url
+        else:
+            public_url = ngrok.connect(port, pyngrok_config=config, bind_tls=True, auth=account).public_url
     except exception.PyngrokNgrokError:
         print(f'Invalid ngrok authtoken, ngrok connection aborted.\n'
               f'Your token: {token}, get the right one on https://dashboard.ngrok.com/get-started/your-authtoken')

modules/paths.py

@@ -9,7 +9,7 @@ sys.path.insert(0, script_path)
 
 # search for directory of stable diffusion in following places
 sd_path = None
-possible_sd_paths = [os.path.join(script_path, 'repositories/stable-diffusion'), '.', os.path.dirname(script_path)]
+possible_sd_paths = [os.path.join(script_path, 'repositories/stable-diffusion-stability-ai'), '.', os.path.dirname(script_path)]
 for possible_sd_path in possible_sd_paths:
     if os.path.exists(os.path.join(possible_sd_path, 'ldm/models/diffusion/ddpm.py')):
         sd_path = os.path.abspath(possible_sd_path)

modules/processing.py

@@ -2,6 +2,7 @@ import json
 import math
 import os
 import sys
+import warnings
 
 import torch
 import numpy as np
@@ -20,7 +21,10 @@ import modules.face_restoration
 import modules.images as images
 import modules.styles
 import logging
+from ldm.data.util import AddMiDaS
+from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion
 
+from einops import repeat, rearrange
 
 # some of those options should not be changed at all because they would break the model, so I removed them from options.
 opt_C = 4
@@ -66,19 +70,15 @@ def apply_overlay(image, paste_loc, index, overlays):
 
     return image
 
-def get_correct_sampler(p):
-    if isinstance(p, modules.processing.StableDiffusionProcessingTxt2Img):
-        return sd_samplers.samplers
-    elif isinstance(p, modules.processing.StableDiffusionProcessingImg2Img):
-        return sd_samplers.samplers_for_img2img
-    elif isinstance(p, modules.api.processing.StableDiffusionProcessingAPI):
-        return sd_samplers.samplers
 
 class StableDiffusionProcessing():
     """
     The first set of paramaters: sd_models -> do_not_reload_embeddings represent the minimum required to create a StableDiffusionProcessing
     """
-    def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_index: int = 0, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = False, tiling: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = 1.0, override_settings: Dict[str, Any] = None):
+    def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, width: int = 512, height: int = 512, restore_faces: bool = False, tiling: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, ddim_discretize: str = None, s_churn: float = 0.0, s_tmax: float = None, s_tmin: float = 0.0, s_noise: float = 1.0, override_settings: Dict[str, Any] = None, sampler_index: int = None):
+        if sampler_index is not None:
+            print("sampler_index argument for StableDiffusionProcessing does not do anything; use sampler_name", file=sys.stderr)
+
         self.sd_model = sd_model
         self.outpath_samples: str = outpath_samples
         self.outpath_grids: str = outpath_grids
@@ -91,7 +91,7 @@ class StableDiffusionProcessing():
         self.subseed_strength: float = subseed_strength
         self.seed_resize_from_h: int = seed_resize_from_h
         self.seed_resize_from_w: int = seed_resize_from_w
-        self.sampler_index: int = sampler_index
+        self.sampler_name: str = sampler_name
         self.batch_size: int = batch_size
         self.n_iter: int = n_iter
         self.steps: int = steps
@@ -116,6 +116,7 @@ class StableDiffusionProcessing():
         self.s_tmax = s_tmax or float('inf')  # not representable as a standard ui option
         self.s_noise = s_noise or opts.s_noise
         self.override_settings = {k: v for k, v in (override_settings or {}).items() if k not in shared.restricted_opts}
+        self.is_using_inpainting_conditioning = False
 
         if not seed_enable_extras:
             self.subseed = -1
@@ -126,6 +127,7 @@ class StableDiffusionProcessing():
         self.scripts = None
         self.script_args = None
         self.all_prompts = None
+        self.all_negative_prompts = None
         self.all_seeds = None
         self.all_subseeds = None
 
@@ -136,6 +138,8 @@ class StableDiffusionProcessing():
             # Pretty sure we can just make this a 1x1 image since its not going to be used besides its batch size.
             return x.new_zeros(x.shape[0], 5, 1, 1)
 
+        self.is_using_inpainting_conditioning = True
+
         height = height or self.height
         width = width or self.width
 
@@ -149,10 +153,27 @@ class StableDiffusionProcessing():
 
         return image_conditioning
 
-    def img2img_image_conditioning(self, source_image, latent_image, image_mask = None):
-        if self.sampler.conditioning_key not in {'hybrid', 'concat'}:
-            # Dummy zero conditioning if we're not using inpainting model.
-            return latent_image.new_zeros(latent_image.shape[0], 5, 1, 1)
+    def depth2img_image_conditioning(self, source_image):
+        # Use the AddMiDaS helper to Format our source image to suit the MiDaS model
+        transformer = AddMiDaS(model_type="dpt_hybrid")
+        transformed = transformer({"jpg": rearrange(source_image[0], "c h w -> h w c")})
+        midas_in = torch.from_numpy(transformed["midas_in"][None, ...]).to(device=shared.device)
+        midas_in = repeat(midas_in, "1 ... -> n ...", n=self.batch_size)
+
+        conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image))
+        conditioning = torch.nn.functional.interpolate(
+            self.sd_model.depth_model(midas_in),
+            size=conditioning_image.shape[2:],
+            mode="bicubic",
+            align_corners=False,
+        )
+
+        (depth_min, depth_max) = torch.aminmax(conditioning)
+        conditioning = 2. * (conditioning - depth_min) / (depth_max - depth_min) - 1.
+        return conditioning
+
+    def inpainting_image_conditioning(self, source_image, latent_image, image_mask = None):
+        self.is_using_inpainting_conditioning = True
 
         # Handle the different mask inputs
         if image_mask is not None:
@@ -188,6 +209,18 @@ class StableDiffusionProcessing():
 
         return image_conditioning
 
+    def img2img_image_conditioning(self, source_image, latent_image, image_mask=None):
+        # HACK: Using introspection as the Depth2Image model doesn't appear to uniquely
+        # identify itself with a field common to all models. The conditioning_key is also hybrid.
+        if isinstance(self.sd_model, LatentDepth2ImageDiffusion):
+            return self.depth2img_image_conditioning(source_image)
+
+        if self.sampler.conditioning_key in {'hybrid', 'concat'}:
+            return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask)
+
+        # Dummy zero conditioning if we're not using inpainting or depth model.
+        return latent_image.new_zeros(latent_image.shape[0], 5, 1, 1)
+
     def init(self, all_prompts, all_seeds, all_subseeds):
         pass
 
@@ -200,7 +233,7 @@ class StableDiffusionProcessing():
 
 
 class Processed:
-    def __init__(self, p: StableDiffusionProcessing, images_list, seed=-1, info="", subseed=None, all_prompts=None, all_seeds=None, all_subseeds=None, index_of_first_image=0, infotexts=None):
+    def __init__(self, p: StableDiffusionProcessing, images_list, seed=-1, info="", subseed=None, all_prompts=None, all_negative_prompts=None, all_seeds=None, all_subseeds=None, index_of_first_image=0, infotexts=None):
         self.images = images_list
         self.prompt = p.prompt
         self.negative_prompt = p.negative_prompt
@@ -210,8 +243,7 @@ class Processed:
         self.info = info
         self.width = p.width
         self.height = p.height
-        self.sampler_index = p.sampler_index
-        self.sampler = sd_samplers.samplers[p.sampler_index].name
+        self.sampler_name = p.sampler_name
         self.cfg_scale = p.cfg_scale
         self.steps = p.steps
         self.batch_size = p.batch_size
@@ -238,17 +270,20 @@ class Processed:
         self.negative_prompt = self.negative_prompt if type(self.negative_prompt) != list else self.negative_prompt[0]
         self.seed = int(self.seed if type(self.seed) != list else self.seed[0]) if self.seed is not None else -1
         self.subseed = int(self.subseed if type(self.subseed) != list else self.subseed[0]) if self.subseed is not None else -1
+        self.is_using_inpainting_conditioning = p.is_using_inpainting_conditioning
 
-        self.all_prompts = all_prompts or [self.prompt]
-        self.all_seeds = all_seeds or [self.seed]
-        self.all_subseeds = all_subseeds or [self.subseed]
+        self.all_prompts = all_prompts or p.all_prompts or [self.prompt]
+        self.all_negative_prompts = all_negative_prompts or p.all_negative_prompts or [self.negative_prompt]
+        self.all_seeds = all_seeds or p.all_seeds or [self.seed]
+        self.all_subseeds = all_subseeds or p.all_subseeds or [self.subseed]
         self.infotexts = infotexts or [info]
 
     def js(self):
         obj = {
-            "prompt": self.prompt,
+            "prompt": self.all_prompts[0],
             "all_prompts": self.all_prompts,
-            "negative_prompt": self.negative_prompt,
+            "negative_prompt": self.all_negative_prompts[0],
+            "all_negative_prompts": self.all_negative_prompts,
             "seed": self.seed,
             "all_seeds": self.all_seeds,
             "subseed": self.subseed,
@@ -256,8 +291,7 @@ class Processed:
             "subseed_strength": self.subseed_strength,
             "width": self.width,
             "height": self.height,
-            "sampler_index": self.sampler_index,
-            "sampler": self.sampler,
+            "sampler_name": self.sampler_name,
             "cfg_scale": self.cfg_scale,
             "steps": self.steps,
             "batch_size": self.batch_size,
@@ -273,6 +307,7 @@ class Processed:
             "styles": self.styles,
             "job_timestamp": self.job_timestamp,
             "clip_skip": self.clip_skip,
+            "is_using_inpainting_conditioning": self.is_using_inpainting_conditioning,
         }
 
         return json.dumps(obj)
@@ -384,7 +419,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration
 
     generation_params = {
         "Steps": p.steps,
-        "Sampler": get_correct_sampler(p)[p.sampler_index].name,
+        "Sampler": p.sampler_name,
         "CFG scale": p.cfg_scale,
         "Seed": all_seeds[index],
         "Face restoration": (opts.face_restoration_model if p.restore_faces else None),
@@ -399,6 +434,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration
         "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength),
         "Seed resize from": (None if p.seed_resize_from_w == 0 or p.seed_resize_from_h == 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"),
         "Denoising strength": getattr(p, 'denoising_strength', None),
+        "Conditional mask weight": getattr(p, "inpainting_mask_weight", shared.opts.inpainting_mask_weight) if p.is_using_inpainting_conditioning else None,
         "Eta": (None if p.sampler is None or p.sampler.eta == p.sampler.default_eta else p.sampler.eta),
         "Clip skip": None if clip_skip <= 1 else clip_skip,
         "ENSD": None if opts.eta_noise_seed_delta == 0 else opts.eta_noise_seed_delta,
@@ -408,7 +444,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments, iteration
 
     generation_params_text = ", ".join([k if k == v else f'{k}: {generation_parameters_copypaste.quote(v)}' for k, v in generation_params.items() if v is not None])
 
-    negative_prompt_text = "\nNegative prompt: " + p.negative_prompt if p.negative_prompt else ""
+    negative_prompt_text = "\nNegative prompt: " + p.all_negative_prompts[index] if  p.all_negative_prompts[index] else ""
 
     return f"{all_prompts[index]}{negative_prompt_text}\n{generation_params_text}".strip()
 
@@ -418,13 +454,15 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
 
     try:
         for k, v in p.override_settings.items():
-            setattr(opts, k, v) # we don't call onchange for simplicity which makes changing model, hypernet impossible
+            setattr(opts, k, v)  # we don't call onchange for simplicity which makes changing model impossible
+            if k == 'sd_hypernetwork': shared.reload_hypernetworks()  # make onchange call for changing hypernet since it is relatively fast to load on-change, while SD models are not
 
         res = process_images_inner(p)
 
-    finally:
+    finally:  # restore opts to original state
         for k, v in stored_opts.items():
             setattr(opts, k, v)
+            if k == 'sd_hypernetwork': shared.reload_hypernetworks()
 
     return res
 
@@ -437,10 +475,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
     else:
         assert p.prompt is not None
 
-    with open(os.path.join(shared.script_path, "params.txt"), "w", encoding="utf8") as file:
-        processed = Processed(p, [], p.seed, "")
-        file.write(processed.infotext(p, 0))
-
     devices.torch_gc()
 
     seed = get_fixed_seed(p.seed)
@@ -451,12 +485,15 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
 
     comments = {}
 
-    shared.prompt_styles.apply_styles(p)
-
     if type(p.prompt) == list:
-        p.all_prompts = p.prompt
+        p.all_prompts = [shared.prompt_styles.apply_styles_to_prompt(x, p.styles) for x in p.prompt]
+    else:
+        p.all_prompts = p.batch_size * p.n_iter * [shared.prompt_styles.apply_styles_to_prompt(p.prompt, p.styles)]
+
+    if type(p.negative_prompt) == list:
+        p.all_negative_prompts = [shared.prompt_styles.apply_negative_styles_to_prompt(x, p.styles) for x in p.negative_prompt]
     else:
-        p.all_prompts = p.batch_size * p.n_iter * [p.prompt]
+        p.all_negative_prompts = p.batch_size * p.n_iter * [shared.prompt_styles.apply_negative_styles_to_prompt(p.negative_prompt, p.styles)]
 
     if type(seed) == list:
         p.all_seeds = seed
@@ -471,6 +508,10 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
     def infotext(iteration=0, position_in_batch=0):
         return create_infotext(p, p.all_prompts, p.all_seeds, p.all_subseeds, comments, iteration, position_in_batch)
 
+    with open(os.path.join(shared.script_path, "params.txt"), "w", encoding="utf8") as file:
+        processed = Processed(p, [], p.seed, "")
+        file.write(processed.infotext(p, 0))
+
     if os.path.exists(cmd_opts.embeddings_dir) and not p.do_not_reload_embeddings:
         model_hijack.embedding_db.load_textual_inversion_embeddings()
 
@@ -495,6 +536,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
                 break
 
             prompts = p.all_prompts[n * p.batch_size:(n + 1) * p.batch_size]
+            negative_prompts = p.all_negative_prompts[n * p.batch_size:(n + 1) * p.batch_size]
             seeds = p.all_seeds[n * p.batch_size:(n + 1) * p.batch_size]
             subseeds = p.all_subseeds[n * p.batch_size:(n + 1) * p.batch_size]
 
@@ -505,7 +547,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
                 p.scripts.process_batch(p, batch_number=n, prompts=prompts, seeds=seeds, subseeds=subseeds)
 
             with devices.autocast():
-                uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps)
+                uc = prompt_parser.get_learned_conditioning(shared.sd_model, negative_prompts, p.steps)
                 c = prompt_parser.get_multicond_learned_conditioning(shared.sd_model, prompts, p.steps)
 
             if len(model_hijack.comments) > 0:
@@ -518,8 +560,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
             with devices.autocast():
                 samples_ddim = p.sample(conditioning=c, unconditional_conditioning=uc, seeds=seeds, subseeds=subseeds, subseed_strength=p.subseed_strength, prompts=prompts)
 
-            samples_ddim = samples_ddim.to(devices.dtype_vae)
-            x_samples_ddim = decode_first_stage(p.sd_model, samples_ddim)
+            x_samples_ddim = [decode_first_stage(p.sd_model, samples_ddim[i:i+1].to(dtype=devices.dtype_vae))[0].cpu() for i in range(samples_ddim.size(0))]
+            x_samples_ddim = torch.stack(x_samples_ddim).float()
             x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0)
 
             del samples_ddim
@@ -591,7 +633,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
 
     devices.torch_gc()
 
-    res = Processed(p, output_images, p.all_seeds[0], infotext() + "".join(["\n\n" + x for x in comments]), subseed=p.all_subseeds[0], all_prompts=p.all_prompts, all_seeds=p.all_seeds, all_subseeds=p.all_subseeds, index_of_first_image=index_of_first_image, infotexts=infotexts)
+    res = Processed(p, output_images, p.all_seeds[0], infotext() + "".join(["\n\n" + x for x in comments]), subseed=p.all_subseeds[0], index_of_first_image=index_of_first_image, infotexts=infotexts)
 
     if p.scripts is not None:
         p.scripts.postprocess(p, res)
@@ -645,7 +687,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
             self.truncate_y = int(self.firstphase_height - firstphase_height_truncated) // opt_f
 
     def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts):
-        self.sampler = sd_samplers.create_sampler_with_index(sd_samplers.samplers, self.sampler_index, self.sd_model)
+        self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model)
 
         if not self.enable_hr:
             x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self)
@@ -706,7 +748,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
 
         shared.state.nextjob()
 
-        self.sampler = sd_samplers.create_sampler_with_index(sd_samplers.samplers, self.sampler_index, self.sd_model)
+        self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model)
 
         noise = create_random_tensors(samples.shape[1:], seeds=seeds, subseeds=subseeds, subseed_strength=subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self)
 
@@ -722,7 +764,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
 class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
     sampler = None
 
-    def __init__(self, init_images: list=None, resize_mode: int=0, denoising_strength: float=0.75, mask: Any=None, mask_blur: int=4, inpainting_fill: int=0, inpaint_full_res: bool=True, inpaint_full_res_padding: int=0, inpainting_mask_invert: int=0, **kwargs):
+    def __init__(self, init_images: list = None, resize_mode: int = 0, denoising_strength: float = 0.75, mask: Any = None, mask_blur: int = 4, inpainting_fill: int = 0, inpaint_full_res: bool = True, inpaint_full_res_padding: int = 0, inpainting_mask_invert: int = 0, initial_noise_multiplier: float = None, **kwargs):
         super().__init__(**kwargs)
 
         self.init_images = init_images
@@ -730,7 +772,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
         self.denoising_strength: float = denoising_strength
         self.init_latent = None
         self.image_mask = mask
-        #self.image_unblurred_mask = None
         self.latent_mask = None
         self.mask_for_overlay = None
         self.mask_blur = mask_blur
@@ -738,44 +779,45 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
         self.inpaint_full_res = inpaint_full_res
         self.inpaint_full_res_padding = inpaint_full_res_padding
         self.inpainting_mask_invert = inpainting_mask_invert
+        self.initial_noise_multiplier = opts.initial_noise_multiplier if initial_noise_multiplier is None else initial_noise_multiplier
         self.mask = None
         self.nmask = None
         self.image_conditioning = None
 
     def init(self, all_prompts, all_seeds, all_subseeds):
-        self.sampler = sd_samplers.create_sampler_with_index(sd_samplers.samplers_for_img2img, self.sampler_index, self.sd_model)
+        self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model)
         crop_region = None
 
-        if self.image_mask is not None:
-            self.image_mask = self.image_mask.convert('L')
+        image_mask = self.image_mask
 
-            if self.inpainting_mask_invert:
-                self.image_mask = ImageOps.invert(self.image_mask)
+        if image_mask is not None:
+            image_mask = image_mask.convert('L')
 
-            #self.image_unblurred_mask = self.image_mask
+            if self.inpainting_mask_invert:
+                image_mask = ImageOps.invert(image_mask)
 
             if self.mask_blur > 0:
-                self.image_mask = self.image_mask.filter(ImageFilter.GaussianBlur(self.mask_blur))
+                image_mask = image_mask.filter(ImageFilter.GaussianBlur(self.mask_blur))
 
             if self.inpaint_full_res:
-                self.mask_for_overlay = self.image_mask
-                mask = self.image_mask.convert('L')
+                self.mask_for_overlay = image_mask
+                mask = image_mask.convert('L')
                 crop_region = masking.get_crop_region(np.array(mask), self.inpaint_full_res_padding)
                 crop_region = masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height)
                 x1, y1, x2, y2 = crop_region
 
                 mask = mask.crop(crop_region)
-                self.image_mask = images.resize_image(2, mask, self.width, self.height)
+                image_mask = images.resize_image(2, mask, self.width, self.height)
                 self.paste_to = (x1, y1, x2-x1, y2-y1)
             else:
-                self.image_mask = images.resize_image(self.resize_mode, self.image_mask, self.width, self.height)
-                np_mask = np.array(self.image_mask)
+                image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height)
+                np_mask = np.array(image_mask)
                 np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8)
                 self.mask_for_overlay = Image.fromarray(np_mask)
 
             self.overlay_images = []
 
-        latent_mask = self.latent_mask if self.latent_mask is not None else self.image_mask
+        latent_mask = self.latent_mask if self.latent_mask is not None else image_mask
 
         add_color_corrections = opts.img2img_color_correction and self.color_corrections is None
         if add_color_corrections:
@@ -787,7 +829,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
             if crop_region is None:
                 image = images.resize_image(self.resize_mode, image, self.width, self.height)
 
-            if self.image_mask is not None:
+            if image_mask is not None:
                 image_masked = Image.new('RGBa', (image.width, image.height))
                 image_masked.paste(image.convert("RGBA").convert("RGBa"), mask=ImageOps.invert(self.mask_for_overlay.convert('L')))
 
@@ -797,7 +839,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
                 image = image.crop(crop_region)
                 image = images.resize_image(2, image, self.width, self.height)
 
-            if self.image_mask is not None:
+            if image_mask is not None:
                 if self.inpainting_fill != 1:
                     image = masking.fill(image, latent_mask)
 
@@ -829,7 +871,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
 
         self.init_latent = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(image))
 
-        if self.image_mask is not None:
+        if image_mask is not None:
             init_mask = latent_mask
             latmask = init_mask.convert('RGB').resize((self.init_latent.shape[3], self.init_latent.shape[2]))
             latmask = np.moveaxis(np.array(latmask, dtype=np.float32), 2, 0) / 255
@@ -846,11 +888,15 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
             elif self.inpainting_fill == 3:
                 self.init_latent = self.init_latent * self.mask
 
-        self.image_conditioning = self.img2img_image_conditioning(image, self.init_latent, self.image_mask)
+        self.image_conditioning = self.img2img_image_conditioning(image, self.init_latent, image_mask)
 
     def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts):
         x = create_random_tensors([opt_C, self.height // opt_f, self.width // opt_f], seeds=seeds, subseeds=subseeds, subseed_strength=self.subseed_strength, seed_resize_from_h=self.seed_resize_from_h, seed_resize_from_w=self.seed_resize_from_w, p=self)
 
+        if self.initial_noise_multiplier != 1.0:
+            self.extra_generation_params["Noise multiplier"] = self.initial_noise_multiplier
+            x *= self.initial_noise_multiplier
+
         samples = self.sampler.sample_img2img(self, self.init_latent, x, conditioning, unconditional_conditioning, image_conditioning=self.image_conditioning)
 
         if self.mask is not None:

modules/safe.py

@@ -62,14 +62,12 @@ class RestrictedUnpickler(pickle.Unpickler):
         raise Exception(f"global '{module}/{name}' is forbidden")
 
 
-allowed_zip_names = ["archive/data.pkl", "archive/version"]
-allowed_zip_names_re = re.compile(r"^archive/data/\d+$")
-
+# Regular expression that accepts 'dirname/version', 'dirname/data.pkl', and 'dirname/data/<number>'
+allowed_zip_names_re = re.compile(r"^([^/]+)/((data/\d+)|version|(data\.pkl))$")
+data_pkl_re = re.compile(r"^([^/]+)/data\.pkl$")
 
 def check_zip_filenames(filename, names):
     for name in names:
-        if name in allowed_zip_names:
-            continue
         if allowed_zip_names_re.match(name):
             continue
 
@@ -82,8 +80,14 @@ def check_pt(filename, extra_handler):
         # new pytorch format is a zip file
         with zipfile.ZipFile(filename) as z:
             check_zip_filenames(filename, z.namelist())
-
-            with z.open('archive/data.pkl') as file:
+            
+            # find filename of data.pkl in zip file: '<directory name>/data.pkl'
+            data_pkl_filenames = [f for f in z.namelist() if data_pkl_re.match(f)]
+            if len(data_pkl_filenames) == 0:
+                raise Exception(f"data.pkl not found in {filename}")
+            if len(data_pkl_filenames) > 1:
+                raise Exception(f"Multiple data.pkl found in {filename}")
+            with z.open(data_pkl_filenames[0]) as file:
                 unpickler = RestrictedUnpickler(file)
                 unpickler.extra_handler = extra_handler
                 unpickler.load()

modules/script_callbacks.py

@@ -7,6 +7,7 @@ from typing import Optional
 from fastapi import FastAPI
 from gradio import Blocks
 
+
 def report_exception(c, job):
     print(f"Error executing callback {job} for {c.script}", file=sys.stderr)
     print(traceback.format_exc(), file=sys.stderr)
@@ -45,15 +46,23 @@ class CFGDenoiserParams:
         """Total number of sampling steps planned"""
 
 
+class UiTrainTabParams:
+    def __init__(self, txt2img_preview_params):
+        self.txt2img_preview_params = txt2img_preview_params
+
+
 ScriptCallback = namedtuple("ScriptCallback", ["script", "callback"])
 callback_map = dict(
     callbacks_app_started=[],
     callbacks_model_loaded=[],
     callbacks_ui_tabs=[],
+    callbacks_ui_train_tabs=[],
     callbacks_ui_settings=[],
     callbacks_before_image_saved=[],
     callbacks_image_saved=[],
-    callbacks_cfg_denoiser=[]
+    callbacks_cfg_denoiser=[],
+    callbacks_before_component=[],
+    callbacks_after_component=[],
 )
 
 
@@ -61,6 +70,7 @@ def clear_callbacks():
     for callback_list in callback_map.values():
         callback_list.clear()
 
+
 def app_started_callback(demo: Optional[Blocks], app: FastAPI):
     for c in callback_map['callbacks_app_started']:
         try:
@@ -79,7 +89,7 @@ def model_loaded_callback(sd_model):
 
 def ui_tabs_callback():
     res = []
-    
+
     for c in callback_map['callbacks_ui_tabs']:
         try:
             res += c.callback() or []
@@ -89,6 +99,14 @@ def ui_tabs_callback():
     return res
 
 
+def ui_train_tabs_callback(params: UiTrainTabParams):
+    for c in callback_map['callbacks_ui_train_tabs']:
+        try:
+            c.callback(params)
+        except Exception:
+            report_exception(c, 'callbacks_ui_train_tabs')
+
+
 def ui_settings_callback():
     for c in callback_map['callbacks_ui_settings']:
         try:
@@ -121,6 +139,22 @@ def cfg_denoiser_callback(params: CFGDenoiserParams):
             report_exception(c, 'cfg_denoiser_callback')
 
 
+def before_component_callback(component, **kwargs):
+    for c in callback_map['callbacks_before_component']:
+        try:
+            c.callback(component, **kwargs)
+        except Exception:
+            report_exception(c, 'before_component_callback')
+
+
+def after_component_callback(component, **kwargs):
+    for c in callback_map['callbacks_after_component']:
+        try:
+            c.callback(component, **kwargs)
+        except Exception:
+            report_exception(c, 'after_component_callback')
+
+
 def add_callback(callbacks, fun):
     stack = [x for x in inspect.stack() if x.filename != __file__]
     filename = stack[0].filename if len(stack) > 0 else 'unknown file'
@@ -169,6 +203,13 @@ def on_ui_tabs(callback):
     add_callback(callback_map['callbacks_ui_tabs'], callback)
 
 
+def on_ui_train_tabs(callback):
+    """register a function to be called when the UI is creating new tabs for the train tab.
+    Create your new tabs with gr.Tab.
+    """
+    add_callback(callback_map['callbacks_ui_train_tabs'], callback)
+
+
 def on_ui_settings(callback):
     """register a function to be called before UI settings are populated; add your settings
     by using shared.opts.add_option(shared.OptionInfo(...)) """
@@ -197,3 +238,20 @@ def on_cfg_denoiser(callback):
         - params: CFGDenoiserParams - parameters to be passed to the inner model and sampling state details.
     """
     add_callback(callback_map['callbacks_cfg_denoiser'], callback)
+
+
+def on_before_component(callback):
+    """register a function to be called before a component is created.
+    The callback is called with arguments:
+        - component - gradio component that is about to be created.
+        - **kwargs - args to gradio.components.IOComponent.__init__ function
+
+    Use elem_id/label fields of kwargs to figure out which component it is.
+    This can be useful to inject your own components somewhere in the middle of vanilla UI.
+    """
+    add_callback(callback_map['callbacks_before_component'], callback)
+
+
+def on_after_component(callback):
+    """register a function to be called after a component is created. See on_before_component for more."""
+    add_callback(callback_map['callbacks_after_component'], callback)

modules/script_loading.py

@@ -0,0 +1,34 @@
+import os
+import sys
+import traceback
+from types import ModuleType
+
+
+def load_module(path):
+    with open(path, "r", encoding="utf8") as file:
+        text = file.read()
+
+    compiled = compile(text, path, 'exec')
+    module = ModuleType(os.path.basename(path))
+    exec(compiled, module.__dict__)
+
+    return module
+
+
+def preload_extensions(extensions_dir, parser):
+    if not os.path.isdir(extensions_dir):
+        return
+
+    for dirname in sorted(os.listdir(extensions_dir)):
+        preload_script = os.path.join(extensions_dir, dirname, "preload.py")
+        if not os.path.isfile(preload_script):
+            continue
+
+        try:
+            module = load_module(preload_script)
+            if hasattr(module, 'preload'):
+                module.preload(parser)
+
+        except Exception:
+            print(f"Error running preload() for {preload_script}", file=sys.stderr)
+            print(traceback.format_exc(), file=sys.stderr)

modules/scripts.py

@@ -6,7 +6,7 @@ from collections import namedtuple
 import gradio as gr
 
 from modules.processing import StableDiffusionProcessing
-from modules import shared, paths, script_callbacks, extensions
+from modules import shared, paths, script_callbacks, extensions, script_loading
 
 AlwaysVisible = object()
 
@@ -17,6 +17,9 @@ class Script:
     args_to = None
     alwayson = False
 
+    is_txt2img = False
+    is_img2img = False
+
     """A gr.Group component that has all script's UI inside it"""
     group = None
 
@@ -93,6 +96,23 @@ class Script:
 
         pass
 
+    def before_component(self, component, **kwargs):
+        """
+        Called before a component is created.
+        Use elem_id/label fields of kwargs to figure out which component it is.
+        This can be useful to inject your own components somewhere in the middle of vanilla UI.
+        You can return created components in the ui() function to add them to the list of arguments for your processing functions
+        """
+
+        pass
+
+    def after_component(self, component, **kwargs):
+        """
+        Called after a component is created. Same as above.
+        """
+
+        pass
+
     def describe(self):
         """unused"""
         return ""
@@ -140,7 +160,7 @@ def list_files_with_name(filename):
             continue
 
         path = os.path.join(dirpath, filename)
-        if os.path.isfile(filename):
+        if os.path.isfile(path):
             res.append(path)
 
     return res
@@ -161,13 +181,7 @@ def load_scripts():
                 sys.path = [scriptfile.basedir] + sys.path
             current_basedir = scriptfile.basedir
 
-            with open(scriptfile.path, "r", encoding="utf8") as file:
-                text = file.read()
-
-            from types import ModuleType
-            compiled = compile(text, scriptfile.path, 'exec')
-            module = ModuleType(scriptfile.filename)
-            exec(compiled, module.__dict__)
+            module = script_loading.load_module(scriptfile.path)
 
             for key, script_class in module.__dict__.items():
                 if type(script_class) == type and issubclass(script_class, Script):
@@ -201,12 +215,18 @@ class ScriptRunner:
         self.titles = []
         self.infotext_fields = []
 
-    def setup_ui(self, is_img2img):
+    def initialize_scripts(self, is_img2img):
+        self.scripts.clear()
+        self.alwayson_scripts.clear()
+        self.selectable_scripts.clear()
+
         for script_class, path, basedir in scripts_data:
             script = script_class()
             script.filename = path
+            script.is_txt2img = not is_img2img
+            script.is_img2img = is_img2img
 
-            visibility = script.show(is_img2img)
+            visibility = script.show(script.is_img2img)
 
             if visibility == AlwaysVisible:
                 self.scripts.append(script)
@@ -217,6 +237,7 @@ class ScriptRunner:
                 self.scripts.append(script)
                 self.selectable_scripts.append(script)
 
+    def setup_ui(self):
         self.titles = [wrap_call(script.title, script.filename, "title") or f"{script.filename} [error]" for script in self.selectable_scripts]
 
         inputs = [None]
@@ -226,7 +247,7 @@ class ScriptRunner:
             script.args_from = len(inputs)
             script.args_to = len(inputs)
 
-            controls = wrap_call(script.ui, script.filename, "ui", is_img2img)
+            controls = wrap_call(script.ui, script.filename, "ui", script.is_img2img)
 
             if controls is None:
                 return
@@ -326,33 +347,44 @@ class ScriptRunner:
                 print(f"Error running postprocess: {script.filename}", file=sys.stderr)
                 print(traceback.format_exc(), file=sys.stderr)
 
+    def before_component(self, component, **kwargs):
+        for script in self.scripts:
+            try:
+                script.before_component(component, **kwargs)
+            except Exception:
+                print(f"Error running before_component: {script.filename}", file=sys.stderr)
+                print(traceback.format_exc(), file=sys.stderr)
+
+    def after_component(self, component, **kwargs):
+        for script in self.scripts:
+            try:
+                script.after_component(component, **kwargs)
+            except Exception:
+                print(f"Error running after_component: {script.filename}", file=sys.stderr)
+                print(traceback.format_exc(), file=sys.stderr)
+
     def reload_sources(self, cache):
         for si, script in list(enumerate(self.scripts)):
-            with open(script.filename, "r", encoding="utf8") as file:
-                args_from = script.args_from
-                args_to = script.args_to
-                filename = script.filename
-                text = file.read()
+            args_from = script.args_from
+            args_to = script.args_to
+            filename = script.filename
 
-                from types import ModuleType
+            module = cache.get(filename, None)
+            if module is None:
+                module = script_loading.load_module(script.filename)
+                cache[filename] = module
 
-                module = cache.get(filename, None)
-                if module is None:
-                    compiled = compile(text, filename, 'exec')
-                    module = ModuleType(script.filename)
-                    exec(compiled, module.__dict__)
-                    cache[filename] = module
-
-                for key, script_class in module.__dict__.items():
-                    if type(script_class) == type and issubclass(script_class, Script):
-                        self.scripts[si] = script_class()
-                        self.scripts[si].filename = filename
-                        self.scripts[si].args_from = args_from
-                        self.scripts[si].args_to = args_to
+            for key, script_class in module.__dict__.items():
+                if type(script_class) == type and issubclass(script_class, Script):
+                    self.scripts[si] = script_class()
+                    self.scripts[si].filename = filename
+                    self.scripts[si].args_from = args_from
+                    self.scripts[si].args_to = args_to
 
 
 scripts_txt2img = ScriptRunner()
 scripts_img2img = ScriptRunner()
+scripts_current: ScriptRunner = None
 
 
 def reload_script_body_only():
@@ -369,3 +401,22 @@ def reload_scripts():
     scripts_txt2img = ScriptRunner()
     scripts_img2img = ScriptRunner()
 
+
+def IOComponent_init(self, *args, **kwargs):
+    if scripts_current is not None:
+        scripts_current.before_component(self, **kwargs)
+
+    script_callbacks.before_component_callback(self, **kwargs)
+
+    res = original_IOComponent_init(self, *args, **kwargs)
+
+    script_callbacks.after_component_callback(self, **kwargs)
+
+    if scripts_current is not None:
+        scripts_current.after_component(self, **kwargs)
+
+    return res
+
+
+original_IOComponent_init = gr.components.IOComponent.__init__
+gr.components.IOComponent.__init__ = IOComponent_init

modules/sd_hijack.py

@@ -1,29 +1,40 @@
-import math
-import os
-import sys
-import traceback
 import torch
-import numpy as np
-from torch import einsum
 from torch.nn.functional import silu
 
 import modules.textual_inversion.textual_inversion
-from modules import prompt_parser, devices, sd_hijack_optimizations, shared
-from modules.shared import opts, device, cmd_opts
+from modules import devices, sd_hijack_optimizations, shared, sd_hijack_checkpoint
+from modules.hypernetworks import hypernetwork
+from modules.shared import cmd_opts
+from modules import sd_hijack_clip, sd_hijack_open_clip, sd_hijack_unet
+
 from modules.sd_hijack_optimizations import invokeAI_mps_available
 
 import ldm.modules.attention
 import ldm.modules.diffusionmodules.model
+import ldm.modules.diffusionmodules.openaimodel
+import ldm.models.diffusion.ddim
+import ldm.models.diffusion.plms
+import ldm.modules.encoders.modules
 
 attention_CrossAttention_forward = ldm.modules.attention.CrossAttention.forward
 diffusionmodules_model_nonlinearity = ldm.modules.diffusionmodules.model.nonlinearity
 diffusionmodules_model_AttnBlock_forward = ldm.modules.diffusionmodules.model.AttnBlock.forward
 
+# new memory efficient cross attention blocks do not support hypernets and we already
+# have memory efficient cross attention anyway, so this disables SD2.0's memory efficient cross attention
+ldm.modules.attention.MemoryEfficientCrossAttention = ldm.modules.attention.CrossAttention
+ldm.modules.attention.BasicTransformerBlock.ATTENTION_MODES["softmax-xformers"] = ldm.modules.attention.CrossAttention
+
+# silence new console spam from SD2
+ldm.modules.attention.print = lambda *args: None
+ldm.modules.diffusionmodules.model.print = lambda *args: None
+
 
 def apply_optimizations():
     undo_optimizations()
 
     ldm.modules.diffusionmodules.model.nonlinearity = silu
+    ldm.modules.diffusionmodules.openaimodel.th = sd_hijack_unet.th
 
     if cmd_opts.force_enable_xformers or (cmd_opts.xformers and shared.xformers_available and torch.version.cuda and (6, 0) <= torch.cuda.get_device_capability(shared.device) <= (9, 0)):
         print("Applying xformers cross attention optimization.")
@@ -47,16 +58,15 @@ def apply_optimizations():
 
 
 def undo_optimizations():
-    from modules.hypernetworks import hypernetwork
-
     ldm.modules.attention.CrossAttention.forward = hypernetwork.attention_CrossAttention_forward
     ldm.modules.diffusionmodules.model.nonlinearity = diffusionmodules_model_nonlinearity
     ldm.modules.diffusionmodules.model.AttnBlock.forward = diffusionmodules_model_AttnBlock_forward
 
 
-def get_target_prompt_token_count(token_count):
-    return math.ceil(max(token_count, 1) / 75) * 75
-
+def fix_checkpoint():
+    ldm.modules.attention.BasicTransformerBlock.forward = sd_hijack_checkpoint.BasicTransformerBlock_forward
+    ldm.modules.diffusionmodules.openaimodel.ResBlock.forward = sd_hijack_checkpoint.ResBlock_forward
+    ldm.modules.diffusionmodules.openaimodel.AttentionBlock.forward = sd_hijack_checkpoint.AttentionBlock_forward
 
 class StableDiffusionModelHijack:
     fixes = None
@@ -68,14 +78,18 @@ class StableDiffusionModelHijack:
     embedding_db = modules.textual_inversion.textual_inversion.EmbeddingDatabase(cmd_opts.embeddings_dir)
 
     def hijack(self, m):
-        model_embeddings = m.cond_stage_model.transformer.text_model.embeddings
-
-        model_embeddings.token_embedding = EmbeddingsWithFixes(model_embeddings.token_embedding, self)
-        m.cond_stage_model = FrozenCLIPEmbedderWithCustomWords(m.cond_stage_model, self)
+        if type(m.cond_stage_model) == ldm.modules.encoders.modules.FrozenCLIPEmbedder:
+            model_embeddings = m.cond_stage_model.transformer.text_model.embeddings
+            model_embeddings.token_embedding = EmbeddingsWithFixes(model_embeddings.token_embedding, self)
+            m.cond_stage_model = sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords(m.cond_stage_model, self)
+        elif type(m.cond_stage_model) == ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder:
+            m.cond_stage_model.model.token_embedding = EmbeddingsWithFixes(m.cond_stage_model.model.token_embedding, self)
+            m.cond_stage_model = sd_hijack_open_clip.FrozenOpenCLIPEmbedderWithCustomWords(m.cond_stage_model, self)
 
         self.clip = m.cond_stage_model
 
         apply_optimizations()
+        fix_checkpoint()
 
         def flatten(el):
             flattened = [flatten(children) for children in el.children()]
@@ -87,15 +101,18 @@ class StableDiffusionModelHijack:
         self.layers = flatten(m)
 
     def undo_hijack(self, m):
-        if type(m.cond_stage_model) == FrozenCLIPEmbedderWithCustomWords:
+        if type(m.cond_stage_model) == sd_hijack_clip.FrozenCLIPEmbedderWithCustomWords:
             m.cond_stage_model = m.cond_stage_model.wrapped
 
-        model_embeddings = m.cond_stage_model.transformer.text_model.embeddings
-        if type(model_embeddings.token_embedding) == EmbeddingsWithFixes:
-            model_embeddings.token_embedding = model_embeddings.token_embedding.wrapped
+            model_embeddings = m.cond_stage_model.transformer.text_model.embeddings
+            if type(model_embeddings.token_embedding) == EmbeddingsWithFixes:
+                model_embeddings.token_embedding = model_embeddings.token_embedding.wrapped
+        elif type(m.cond_stage_model) == sd_hijack_open_clip.FrozenOpenCLIPEmbedderWithCustomWords:
+            m.cond_stage_model.wrapped.model.token_embedding = m.cond_stage_model.wrapped.model.token_embedding.wrapped
+            m.cond_stage_model = m.cond_stage_model.wrapped
 
+        self.apply_circular(False)
         self.layers = None
-        self.circular_enabled = False
         self.clip = None
 
     def apply_circular(self, enable):
@@ -112,262 +129,9 @@ class StableDiffusionModelHijack:
 
     def tokenize(self, text):
         _, remade_batch_tokens, _, _, _, token_count = self.clip.process_text([text])
-        return remade_batch_tokens[0], token_count, get_target_prompt_token_count(token_count)
+        return remade_batch_tokens[0], token_count, sd_hijack_clip.get_target_prompt_token_count(token_count)
 
 
-class FrozenCLIPEmbedderWithCustomWords(torch.nn.Module):
-    def __init__(self, wrapped, hijack):
-        super().__init__()
-        self.wrapped = wrapped
-        self.hijack: StableDiffusionModelHijack = hijack
-        self.tokenizer = wrapped.tokenizer
-        self.token_mults = {}
-
-        self.comma_token = [v for k, v in self.tokenizer.get_vocab().items() if k == ',</w>'][0]
-
-        tokens_with_parens = [(k, v) for k, v in self.tokenizer.get_vocab().items() if '(' in k or ')' in k or '[' in k or ']' in k]
-        for text, ident in tokens_with_parens:
-            mult = 1.0
-            for c in text:
-                if c == '[':
-                    mult /= 1.1
-                if c == ']':
-                    mult *= 1.1
-                if c == '(':
-                    mult *= 1.1
-                if c == ')':
-                    mult /= 1.1
-
-            if mult != 1.0:
-                self.token_mults[ident] = mult
-
-    def tokenize_line(self, line, used_custom_terms, hijack_comments):
-        id_end = self.wrapped.tokenizer.eos_token_id
-
-        if opts.enable_emphasis:
-            parsed = prompt_parser.parse_prompt_attention(line)
-        else:
-            parsed = [[line, 1.0]]
-
-        tokenized = self.wrapped.tokenizer([text for text, _ in parsed], truncation=False, add_special_tokens=False)["input_ids"]
-
-        fixes = []
-        remade_tokens = []
-        multipliers = []
-        last_comma = -1
-
-        for tokens, (text, weight) in zip(tokenized, parsed):
-            i = 0
-            while i < len(tokens):
-                token = tokens[i]
-
-                embedding, embedding_length_in_tokens = self.hijack.embedding_db.find_embedding_at_position(tokens, i)
-
-                if token == self.comma_token:
-                    last_comma = len(remade_tokens)
-                elif opts.comma_padding_backtrack != 0 and max(len(remade_tokens), 1) % 75 == 0 and last_comma != -1 and len(remade_tokens) - last_comma <= opts.comma_padding_backtrack:
-                    last_comma += 1
-                    reloc_tokens = remade_tokens[last_comma:]
-                    reloc_mults = multipliers[last_comma:]
-
-                    remade_tokens = remade_tokens[:last_comma]
-                    length = len(remade_tokens)
-
-                    rem = int(math.ceil(length / 75)) * 75 - length
-                    remade_tokens += [id_end] * rem + reloc_tokens
-                    multipliers = multipliers[:last_comma] + [1.0] * rem + reloc_mults
-
-                if embedding is None:
-                    remade_tokens.append(token)
-                    multipliers.append(weight)
-                    i += 1
-                else:
-                    emb_len = int(embedding.vec.shape[0])
-                    iteration = len(remade_tokens) // 75
-                    if (len(remade_tokens) + emb_len) // 75 != iteration:
-                        rem = (75 * (iteration + 1) - len(remade_tokens))
-                        remade_tokens += [id_end] * rem
-                        multipliers += [1.0] * rem
-                        iteration += 1
-                    fixes.append((iteration, (len(remade_tokens) % 75, embedding)))
-                    remade_tokens += [0] * emb_len
-                    multipliers += [weight] * emb_len
-                    used_custom_terms.append((embedding.name, embedding.checksum()))
-                    i += embedding_length_in_tokens
-
-        token_count = len(remade_tokens)
-        prompt_target_length = get_target_prompt_token_count(token_count)
-        tokens_to_add = prompt_target_length - len(remade_tokens)
-
-        remade_tokens = remade_tokens + [id_end] * tokens_to_add
-        multipliers = multipliers + [1.0] * tokens_to_add
-
-        return remade_tokens, fixes, multipliers, token_count
-
-    def process_text(self, texts):
-        used_custom_terms = []
-        remade_batch_tokens = []
-        hijack_comments = []
-        hijack_fixes = []
-        token_count = 0
-
-        cache = {}
-        batch_multipliers = []
-        for line in texts:
-            if line in cache:
-                remade_tokens, fixes, multipliers = cache[line]
-            else:
-                remade_tokens, fixes, multipliers, current_token_count = self.tokenize_line(line, used_custom_terms, hijack_comments)
-                token_count = max(current_token_count, token_count)
-
-                cache[line] = (remade_tokens, fixes, multipliers)
-
-            remade_batch_tokens.append(remade_tokens)
-            hijack_fixes.append(fixes)
-            batch_multipliers.append(multipliers)
-
-        return batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count
-
-    def process_text_old(self, text):
-        id_start = self.wrapped.tokenizer.bos_token_id
-        id_end = self.wrapped.tokenizer.eos_token_id
-        maxlen = self.wrapped.max_length  # you get to stay at 77
-        used_custom_terms = []
-        remade_batch_tokens = []
-        overflowing_words = []
-        hijack_comments = []
-        hijack_fixes = []
-        token_count = 0
-
-        cache = {}
-        batch_tokens = self.wrapped.tokenizer(text, truncation=False, add_special_tokens=False)["input_ids"]
-        batch_multipliers = []
-        for tokens in batch_tokens:
-            tuple_tokens = tuple(tokens)
-
-            if tuple_tokens in cache:
-                remade_tokens, fixes, multipliers = cache[tuple_tokens]
-            else:
-                fixes = []
-                remade_tokens = []
-                multipliers = []
-                mult = 1.0
-
-                i = 0
-                while i < len(tokens):
-                    token = tokens[i]
-
-                    embedding, embedding_length_in_tokens = self.hijack.embedding_db.find_embedding_at_position(tokens, i)
-
-                    mult_change = self.token_mults.get(token) if opts.enable_emphasis else None
-                    if mult_change is not None:
-                        mult *= mult_change
-                        i += 1
-                    elif embedding is None:
-                        remade_tokens.append(token)
-                        multipliers.append(mult)
-                        i += 1
-                    else:
-                        emb_len = int(embedding.vec.shape[0])
-                        fixes.append((len(remade_tokens), embedding))
-                        remade_tokens += [0] * emb_len
-                        multipliers += [mult] * emb_len
-                        used_custom_terms.append((embedding.name, embedding.checksum()))
-                        i += embedding_length_in_tokens
-
-                if len(remade_tokens) > maxlen - 2:
-                    vocab = {v: k for k, v in self.wrapped.tokenizer.get_vocab().items()}
-                    ovf = remade_tokens[maxlen - 2:]
-                    overflowing_words = [vocab.get(int(x), "") for x in ovf]
-                    overflowing_text = self.wrapped.tokenizer.convert_tokens_to_string(''.join(overflowing_words))
-                    hijack_comments.append(f"Warning: too many input tokens; some ({len(overflowing_words)}) have been truncated:\n{overflowing_text}\n")
-
-                token_count = len(remade_tokens)
-                remade_tokens = remade_tokens + [id_end] * (maxlen - 2 - len(remade_tokens))
-                remade_tokens = [id_start] + remade_tokens[0:maxlen - 2] + [id_end]
-                cache[tuple_tokens] = (remade_tokens, fixes, multipliers)
-
-            multipliers = multipliers + [1.0] * (maxlen - 2 - len(multipliers))
-            multipliers = [1.0] + multipliers[0:maxlen - 2] + [1.0]
-
-            remade_batch_tokens.append(remade_tokens)
-            hijack_fixes.append(fixes)
-            batch_multipliers.append(multipliers)
-        return batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count
-
-    def forward(self, text):
-        use_old = opts.use_old_emphasis_implementation
-        if use_old:
-            batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count = self.process_text_old(text)
-        else:
-            batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count = self.process_text(text)
-
-        self.hijack.comments += hijack_comments
-
-        if len(used_custom_terms) > 0:
-            self.hijack.comments.append("Used embeddings: " + ", ".join([f'{word} [{checksum}]' for word, checksum in used_custom_terms]))
-
-        if use_old:
-            self.hijack.fixes = hijack_fixes
-            return self.process_tokens(remade_batch_tokens, batch_multipliers)
-
-        z = None
-        i = 0
-        while max(map(len, remade_batch_tokens)) != 0:
-            rem_tokens = [x[75:] for x in remade_batch_tokens]
-            rem_multipliers = [x[75:] for x in batch_multipliers]
-
-            self.hijack.fixes = []
-            for unfiltered in hijack_fixes:
-                fixes = []
-                for fix in unfiltered:
-                    if fix[0] == i:
-                        fixes.append(fix[1])
-                self.hijack.fixes.append(fixes)
-
-            tokens = []
-            multipliers = []
-            for j in range(len(remade_batch_tokens)):
-                if len(remade_batch_tokens[j]) > 0:
-                    tokens.append(remade_batch_tokens[j][:75])
-                    multipliers.append(batch_multipliers[j][:75])
-                else:
-                    tokens.append([self.wrapped.tokenizer.eos_token_id] * 75)
-                    multipliers.append([1.0] * 75)
-
-            z1 = self.process_tokens(tokens, multipliers)
-            z = z1 if z is None else torch.cat((z, z1), axis=-2)
-
-            remade_batch_tokens = rem_tokens
-            batch_multipliers = rem_multipliers
-            i += 1
-
-        return z
-
-    def process_tokens(self, remade_batch_tokens, batch_multipliers):
-        if not opts.use_old_emphasis_implementation:
-            remade_batch_tokens = [[self.wrapped.tokenizer.bos_token_id] + x[:75] + [self.wrapped.tokenizer.eos_token_id] for x in remade_batch_tokens]
-            batch_multipliers = [[1.0] + x[:75] + [1.0] for x in batch_multipliers]
-
-        tokens = torch.asarray(remade_batch_tokens).to(device)
-        outputs = self.wrapped.transformer(input_ids=tokens, output_hidden_states=-opts.CLIP_stop_at_last_layers)
-
-        if opts.CLIP_stop_at_last_layers > 1:
-            z = outputs.hidden_states[-opts.CLIP_stop_at_last_layers]
-            z = self.wrapped.transformer.text_model.final_layer_norm(z)
-        else:
-            z = outputs.last_hidden_state
-
-        # restoring original mean is likely not correct, but it seems to work well to prevent artifacts that happen otherwise
-        batch_multipliers_of_same_length = [x + [1.0] * (75 - len(x)) for x in batch_multipliers]
-        batch_multipliers = torch.asarray(batch_multipliers_of_same_length).to(device)
-        original_mean = z.mean()
-        z *= batch_multipliers.reshape(batch_multipliers.shape + (1,)).expand(z.shape)
-        new_mean = z.mean()
-        z *= original_mean / new_mean
-
-        return z
-
 
 class EmbeddingsWithFixes(torch.nn.Module):
     def __init__(self, wrapped, embeddings):
@@ -406,3 +170,19 @@ def add_circular_option_to_conv_2d():
 
 
 model_hijack = StableDiffusionModelHijack()
+
+
+def register_buffer(self, name, attr):
+    """
+    Fix register buffer bug for Mac OS.
+    """
+
+    if type(attr) == torch.Tensor:
+        if attr.device != devices.device:
+            attr = attr.to(device=devices.device, dtype=(torch.float32 if devices.device.type == 'mps' else None))
+
+    setattr(self, name, attr)
+
+
+ldm.models.diffusion.ddim.DDIMSampler.register_buffer = register_buffer
+ldm.models.diffusion.plms.PLMSSampler.register_buffer = register_buffer

modules/sd_hijack_checkpoint.py

@@ -0,0 +1,10 @@
+from torch.utils.checkpoint import checkpoint
+
+def BasicTransformerBlock_forward(self, x, context=None):
+    return checkpoint(self._forward, x, context)
+
+def AttentionBlock_forward(self, x):
+    return checkpoint(self._forward, x)
+
+def ResBlock_forward(self, x, emb):
+    return checkpoint(self._forward, x, emb)

modules/sd_hijack_clip.py

@@ -0,0 +1,301 @@
+import math
+
+import torch
+
+from modules import prompt_parser, devices
+from modules.shared import opts
+
+
+def get_target_prompt_token_count(token_count):
+    return math.ceil(max(token_count, 1) / 75) * 75
+
+
+class FrozenCLIPEmbedderWithCustomWordsBase(torch.nn.Module):
+    def __init__(self, wrapped, hijack):
+        super().__init__()
+        self.wrapped = wrapped
+        self.hijack = hijack
+
+    def tokenize(self, texts):
+        raise NotImplementedError
+
+    def encode_with_transformers(self, tokens):
+        raise NotImplementedError
+
+    def encode_embedding_init_text(self, init_text, nvpt):
+        raise NotImplementedError
+
+    def tokenize_line(self, line, used_custom_terms, hijack_comments):
+        if opts.enable_emphasis:
+            parsed = prompt_parser.parse_prompt_attention(line)
+        else:
+            parsed = [[line, 1.0]]
+
+        tokenized = self.tokenize([text for text, _ in parsed])
+
+        fixes = []
+        remade_tokens = []
+        multipliers = []
+        last_comma = -1
+
+        for tokens, (text, weight) in zip(tokenized, parsed):
+            i = 0
+            while i < len(tokens):
+                token = tokens[i]
+
+                embedding, embedding_length_in_tokens = self.hijack.embedding_db.find_embedding_at_position(tokens, i)
+
+                if token == self.comma_token:
+                    last_comma = len(remade_tokens)
+                elif opts.comma_padding_backtrack != 0 and max(len(remade_tokens), 1) % 75 == 0 and last_comma != -1 and len(remade_tokens) - last_comma <= opts.comma_padding_backtrack:
+                    last_comma += 1
+                    reloc_tokens = remade_tokens[last_comma:]
+                    reloc_mults = multipliers[last_comma:]
+
+                    remade_tokens = remade_tokens[:last_comma]
+                    length = len(remade_tokens)
+
+                    rem = int(math.ceil(length / 75)) * 75 - length
+                    remade_tokens += [self.id_end] * rem + reloc_tokens
+                    multipliers = multipliers[:last_comma] + [1.0] * rem + reloc_mults
+
+                if embedding is None:
+                    remade_tokens.append(token)
+                    multipliers.append(weight)
+                    i += 1
+                else:
+                    emb_len = int(embedding.vec.shape[0])
+                    iteration = len(remade_tokens) // 75
+                    if (len(remade_tokens) + emb_len) // 75 != iteration:
+                        rem = (75 * (iteration + 1) - len(remade_tokens))
+                        remade_tokens += [self.id_end] * rem
+                        multipliers += [1.0] * rem
+                        iteration += 1
+                    fixes.append((iteration, (len(remade_tokens) % 75, embedding)))
+                    remade_tokens += [0] * emb_len
+                    multipliers += [weight] * emb_len
+                    used_custom_terms.append((embedding.name, embedding.checksum()))
+                    i += embedding_length_in_tokens
+
+        token_count = len(remade_tokens)
+        prompt_target_length = get_target_prompt_token_count(token_count)
+        tokens_to_add = prompt_target_length - len(remade_tokens)
+
+        remade_tokens = remade_tokens + [self.id_end] * tokens_to_add
+        multipliers = multipliers + [1.0] * tokens_to_add
+
+        return remade_tokens, fixes, multipliers, token_count
+
+    def process_text(self, texts):
+        used_custom_terms = []
+        remade_batch_tokens = []
+        hijack_comments = []
+        hijack_fixes = []
+        token_count = 0
+
+        cache = {}
+        batch_multipliers = []
+        for line in texts:
+            if line in cache:
+                remade_tokens, fixes, multipliers = cache[line]
+            else:
+                remade_tokens, fixes, multipliers, current_token_count = self.tokenize_line(line, used_custom_terms, hijack_comments)
+                token_count = max(current_token_count, token_count)
+
+                cache[line] = (remade_tokens, fixes, multipliers)
+
+            remade_batch_tokens.append(remade_tokens)
+            hijack_fixes.append(fixes)
+            batch_multipliers.append(multipliers)
+
+        return batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count
+
+    def process_text_old(self, texts):
+        id_start = self.id_start
+        id_end = self.id_end
+        maxlen = self.wrapped.max_length  # you get to stay at 77
+        used_custom_terms = []
+        remade_batch_tokens = []
+        hijack_comments = []
+        hijack_fixes = []
+        token_count = 0
+
+        cache = {}
+        batch_tokens = self.tokenize(texts)
+        batch_multipliers = []
+        for tokens in batch_tokens:
+            tuple_tokens = tuple(tokens)
+
+            if tuple_tokens in cache:
+                remade_tokens, fixes, multipliers = cache[tuple_tokens]
+            else:
+                fixes = []
+                remade_tokens = []
+                multipliers = []
+                mult = 1.0
+
+                i = 0
+                while i < len(tokens):
+                    token = tokens[i]
+
+                    embedding, embedding_length_in_tokens = self.hijack.embedding_db.find_embedding_at_position(tokens, i)
+
+                    mult_change = self.token_mults.get(token) if opts.enable_emphasis else None
+                    if mult_change is not None:
+                        mult *= mult_change
+                        i += 1
+                    elif embedding is None:
+                        remade_tokens.append(token)
+                        multipliers.append(mult)
+                        i += 1
+                    else:
+                        emb_len = int(embedding.vec.shape[0])
+                        fixes.append((len(remade_tokens), embedding))
+                        remade_tokens += [0] * emb_len
+                        multipliers += [mult] * emb_len
+                        used_custom_terms.append((embedding.name, embedding.checksum()))
+                        i += embedding_length_in_tokens
+
+                if len(remade_tokens) > maxlen - 2:
+                    vocab = {v: k for k, v in self.wrapped.tokenizer.get_vocab().items()}
+                    ovf = remade_tokens[maxlen - 2:]
+                    overflowing_words = [vocab.get(int(x), "") for x in ovf]
+                    overflowing_text = self.wrapped.tokenizer.convert_tokens_to_string(''.join(overflowing_words))
+                    hijack_comments.append(f"Warning: too many input tokens; some ({len(overflowing_words)}) have been truncated:\n{overflowing_text}\n")
+
+                token_count = len(remade_tokens)
+                remade_tokens = remade_tokens + [id_end] * (maxlen - 2 - len(remade_tokens))
+                remade_tokens = [id_start] + remade_tokens[0:maxlen - 2] + [id_end]
+                cache[tuple_tokens] = (remade_tokens, fixes, multipliers)
+
+            multipliers = multipliers + [1.0] * (maxlen - 2 - len(multipliers))
+            multipliers = [1.0] + multipliers[0:maxlen - 2] + [1.0]
+
+            remade_batch_tokens.append(remade_tokens)
+            hijack_fixes.append(fixes)
+            batch_multipliers.append(multipliers)
+        return batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count
+
+    def forward(self, text):
+        use_old = opts.use_old_emphasis_implementation
+        if use_old:
+            batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count = self.process_text_old(text)
+        else:
+            batch_multipliers, remade_batch_tokens, used_custom_terms, hijack_comments, hijack_fixes, token_count = self.process_text(text)
+
+        self.hijack.comments += hijack_comments
+
+        if len(used_custom_terms) > 0:
+            self.hijack.comments.append("Used embeddings: " + ", ".join([f'{word} [{checksum}]' for word, checksum in used_custom_terms]))
+
+        if use_old:
+            self.hijack.fixes = hijack_fixes
+            return self.process_tokens(remade_batch_tokens, batch_multipliers)
+
+        z = None
+        i = 0
+        while max(map(len, remade_batch_tokens)) != 0:
+            rem_tokens = [x[75:] for x in remade_batch_tokens]
+            rem_multipliers = [x[75:] for x in batch_multipliers]
+
+            self.hijack.fixes = []
+            for unfiltered in hijack_fixes:
+                fixes = []
+                for fix in unfiltered:
+                    if fix[0] == i:
+                        fixes.append(fix[1])
+                self.hijack.fixes.append(fixes)
+
+            tokens = []
+            multipliers = []
+            for j in range(len(remade_batch_tokens)):
+                if len(remade_batch_tokens[j]) > 0:
+                    tokens.append(remade_batch_tokens[j][:75])
+                    multipliers.append(batch_multipliers[j][:75])
+                else:
+                    tokens.append([self.id_end] * 75)
+                    multipliers.append([1.0] * 75)
+
+            z1 = self.process_tokens(tokens, multipliers)
+            z = z1 if z is None else torch.cat((z, z1), axis=-2)
+
+            remade_batch_tokens = rem_tokens
+            batch_multipliers = rem_multipliers
+            i += 1
+
+        return z
+
+    def process_tokens(self, remade_batch_tokens, batch_multipliers):
+        if not opts.use_old_emphasis_implementation:
+            remade_batch_tokens = [[self.id_start] + x[:75] + [self.id_end] for x in remade_batch_tokens]
+            batch_multipliers = [[1.0] + x[:75] + [1.0] for x in batch_multipliers]
+
+        tokens = torch.asarray(remade_batch_tokens).to(devices.device)
+
+        if self.id_end != self.id_pad:
+            for batch_pos in range(len(remade_batch_tokens)):
+                index = remade_batch_tokens[batch_pos].index(self.id_end)
+                tokens[batch_pos, index+1:tokens.shape[1]] = self.id_pad
+
+        z = self.encode_with_transformers(tokens)
+
+        # restoring original mean is likely not correct, but it seems to work well to prevent artifacts that happen otherwise
+        batch_multipliers_of_same_length = [x + [1.0] * (75 - len(x)) for x in batch_multipliers]
+        batch_multipliers = torch.asarray(batch_multipliers_of_same_length).to(devices.device)
+        original_mean = z.mean()
+        z *= batch_multipliers.reshape(batch_multipliers.shape + (1,)).expand(z.shape)
+        new_mean = z.mean()
+        z *= original_mean / new_mean
+
+        return z
+
+
+class FrozenCLIPEmbedderWithCustomWords(FrozenCLIPEmbedderWithCustomWordsBase):
+    def __init__(self, wrapped, hijack):
+        super().__init__(wrapped, hijack)
+        self.tokenizer = wrapped.tokenizer
+        self.comma_token = [v for k, v in self.tokenizer.get_vocab().items() if k == ',</w>'][0]
+
+        self.token_mults = {}
+        tokens_with_parens = [(k, v) for k, v in self.tokenizer.get_vocab().items() if '(' in k or ')' in k or '[' in k or ']' in k]
+        for text, ident in tokens_with_parens:
+            mult = 1.0
+            for c in text:
+                if c == '[':
+                    mult /= 1.1
+                if c == ']':
+                    mult *= 1.1
+                if c == '(':
+                    mult *= 1.1
+                if c == ')':
+                    mult /= 1.1
+
+            if mult != 1.0:
+                self.token_mults[ident] = mult
+
+        self.id_start = self.wrapped.tokenizer.bos_token_id
+        self.id_end = self.wrapped.tokenizer.eos_token_id
+        self.id_pad = self.id_end
+
+    def tokenize(self, texts):
+        tokenized = self.wrapped.tokenizer(texts, truncation=False, add_special_tokens=False)["input_ids"]
+
+        return tokenized
+
+    def encode_with_transformers(self, tokens):
+        outputs = self.wrapped.transformer(input_ids=tokens, output_hidden_states=-opts.CLIP_stop_at_last_layers)
+
+        if opts.CLIP_stop_at_last_layers > 1:
+            z = outputs.hidden_states[-opts.CLIP_stop_at_last_layers]
+            z = self.wrapped.transformer.text_model.final_layer_norm(z)
+        else:
+            z = outputs.last_hidden_state
+
+        return z
+
+    def encode_embedding_init_text(self, init_text, nvpt):
+        embedding_layer = self.wrapped.transformer.text_model.embeddings
+        ids = self.wrapped.tokenizer(init_text, max_length=nvpt, return_tensors="pt", add_special_tokens=False)["input_ids"]
+        embedded = embedding_layer.token_embedding.wrapped(ids.to(devices.device)).squeeze(0)
+
+        return embedded

modules/sd_hijack_inpainting.py

@@ -199,8 +199,8 @@ def sample_plms(self,
 
 @torch.no_grad()
 def p_sample_plms(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
-                    temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
-                    unconditional_guidance_scale=1., unconditional_conditioning=None, old_eps=None, t_next=None):
+                  temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                  unconditional_guidance_scale=1., unconditional_conditioning=None, old_eps=None, t_next=None, dynamic_threshold=None):
     b, *_, device = *x.shape, x.device
 
     def get_model_output(x, t):
@@ -249,6 +249,8 @@ def p_sample_plms(self, x, c, t, index, repeat_noise=False, use_original_steps=F
         pred_x0 = (x - sqrt_one_minus_at * e_t) / a_t.sqrt()
         if quantize_denoised:
             pred_x0, _, *_ = self.model.first_stage_model.quantize(pred_x0)
+        if dynamic_threshold is not None:
+            pred_x0 = norm_thresholding(pred_x0, dynamic_threshold)
         # direction pointing to x_t
         dir_xt = (1. - a_prev - sigma_t**2).sqrt() * e_t
         noise = sigma_t * noise_like(x.shape, device, repeat_noise) * temperature
@@ -321,11 +323,16 @@ def should_hijack_inpainting(checkpoint_info):
 
 
 def do_inpainting_hijack():
-    ldm.models.diffusion.ddpm.get_unconditional_conditioning = get_unconditional_conditioning
+    # most of this stuff seems to no longer be needed because it is already included into SD2.0
+    # LatentInpaintDiffusion remains because SD2.0's LatentInpaintDiffusion can't be loaded without specifying a checkpoint
+    # p_sample_plms is needed because PLMS can't work with dicts as conditionings
+    # this file should be cleaned up later if weverything tuens out to work fine
+
+    # ldm.models.diffusion.ddpm.get_unconditional_conditioning = get_unconditional_conditioning
     ldm.models.diffusion.ddpm.LatentInpaintDiffusion = LatentInpaintDiffusion
 
-    ldm.models.diffusion.ddim.DDIMSampler.p_sample_ddim = p_sample_ddim
-    ldm.models.diffusion.ddim.DDIMSampler.sample = sample_ddim
+    # ldm.models.diffusion.ddim.DDIMSampler.p_sample_ddim = p_sample_ddim
+    # ldm.models.diffusion.ddim.DDIMSampler.sample = sample_ddim
 
     ldm.models.diffusion.plms.PLMSSampler.p_sample_plms = p_sample_plms
-    ldm.models.diffusion.plms.PLMSSampler.sample = sample_plms
+    # ldm.models.diffusion.plms.PLMSSampler.sample = sample_plms

modules/sd_hijack_open_clip.py

@@ -0,0 +1,37 @@
+import open_clip.tokenizer
+import torch
+
+from modules import sd_hijack_clip, devices
+from modules.shared import opts
+
+tokenizer = open_clip.tokenizer._tokenizer
+
+
+class FrozenOpenCLIPEmbedderWithCustomWords(sd_hijack_clip.FrozenCLIPEmbedderWithCustomWordsBase):
+    def __init__(self, wrapped, hijack):
+        super().__init__(wrapped, hijack)
+
+        self.comma_token = [v for k, v in tokenizer.encoder.items() if k == ',</w>'][0]
+        self.id_start = tokenizer.encoder["<start_of_text>"]
+        self.id_end = tokenizer.encoder["<end_of_text>"]
+        self.id_pad = 0
+
+    def tokenize(self, texts):
+        assert not opts.use_old_emphasis_implementation, 'Old emphasis implementation not supported for Open Clip'
+
+        tokenized = [tokenizer.encode(text) for text in texts]
+
+        return tokenized
+
+    def encode_with_transformers(self, tokens):
+        # set self.wrapped.layer_idx here according to opts.CLIP_stop_at_last_layers
+        z = self.wrapped.encode_with_transformer(tokens)
+
+        return z
+
+    def encode_embedding_init_text(self, init_text, nvpt):
+        ids = tokenizer.encode(init_text)
+        ids = torch.asarray([ids], device=devices.device, dtype=torch.int)
+        embedded = self.wrapped.model.token_embedding.wrapped(ids).squeeze(0)
+
+        return embedded

modules/sd_hijack_unet.py

@@ -0,0 +1,30 @@
+import torch
+
+
+class TorchHijackForUnet:
+    """
+    This is torch, but with cat that resizes tensors to appropriate dimensions if they do not match;
+    this makes it possible to create pictures with dimensions that are muliples of 8 rather than 64
+    """
+
+    def __getattr__(self, item):
+        if item == 'cat':
+            return self.cat
+
+        if hasattr(torch, item):
+            return getattr(torch, item)
+
+        raise AttributeError("'{}' object has no attribute '{}'".format(type(self).__name__, item))
+
+    def cat(self, tensors, *args, **kwargs):
+        if len(tensors) == 2:
+            a, b = tensors
+            if a.shape[-2:] != b.shape[-2:]:
+                a = torch.nn.functional.interpolate(a, b.shape[-2:], mode="nearest")
+
+            tensors = (a, b)
+
+        return torch.cat(tensors, *args, **kwargs)
+
+
+th = TorchHijackForUnet()

modules/sd_models.py

@@ -5,7 +5,11 @@ import gc
 from collections import namedtuple
 import torch
 import re
+import safetensors.torch
 from omegaconf import OmegaConf
+from os import mkdir
+from urllib import request
+import ldm.modules.midas as midas
 
 from ldm.util import instantiate_from_config
 
@@ -35,6 +39,7 @@ def setup_model():
         os.makedirs(model_path)
 
     list_models()
+    enable_midas_autodownload()
 
 
 def checkpoint_tiles(): 
@@ -45,7 +50,7 @@ def checkpoint_tiles():
 
 def list_models():
     checkpoints_list.clear()
-    model_list = modelloader.load_models(model_path=model_path, command_path=shared.cmd_opts.ckpt_dir, ext_filter=[".ckpt"])
+    model_list = modelloader.load_models(model_path=model_path, command_path=shared.cmd_opts.ckpt_dir, ext_filter=[".ckpt", ".safetensors"])
 
     def modeltitle(path, shorthash):
         abspath = os.path.abspath(path)
@@ -143,8 +148,8 @@ def transform_checkpoint_dict_key(k):
 
 
 def get_state_dict_from_checkpoint(pl_sd):
-    if "state_dict" in pl_sd:
-        pl_sd = pl_sd["state_dict"]
+    pl_sd = pl_sd.pop("state_dict", pl_sd)
+    pl_sd.pop("state_dict", None)
 
     sd = {}
     for k, v in pl_sd.items():
@@ -159,27 +164,41 @@ def get_state_dict_from_checkpoint(pl_sd):
     return pl_sd
 
 
+def read_state_dict(checkpoint_file, print_global_state=False, map_location=None):
+    _, extension = os.path.splitext(checkpoint_file)
+    if extension.lower() == ".safetensors":
+        pl_sd = safetensors.torch.load_file(checkpoint_file, device=map_location or shared.weight_load_location)
+    else:
+        pl_sd = torch.load(checkpoint_file, map_location=map_location or shared.weight_load_location)
+
+    if print_global_state and "global_step" in pl_sd:
+        print(f"Global Step: {pl_sd['global_step']}")
+
+    sd = get_state_dict_from_checkpoint(pl_sd)
+    return sd
+
+
 def load_model_weights(model, checkpoint_info, vae_file="auto"):
     checkpoint_file = checkpoint_info.filename
     sd_model_hash = checkpoint_info.hash
 
-    if shared.opts.sd_checkpoint_cache > 0 and hasattr(model, "sd_checkpoint_info"):
-        sd_vae.restore_base_vae(model)
-        checkpoints_loaded[model.sd_checkpoint_info] = model.state_dict().copy()
-
-    vae_file = sd_vae.resolve_vae(checkpoint_file, vae_file=vae_file)
+    cache_enabled = shared.opts.sd_checkpoint_cache > 0
 
-    if checkpoint_info not in checkpoints_loaded:
+    if cache_enabled and checkpoint_info in checkpoints_loaded:
+        # use checkpoint cache
+        print(f"Loading weights [{sd_model_hash}] from cache")
+        model.load_state_dict(checkpoints_loaded[checkpoint_info])
+    else:
+        # load from file
         print(f"Loading weights [{sd_model_hash}] from {checkpoint_file}")
 
-        pl_sd = torch.load(checkpoint_file, map_location=shared.weight_load_location)
-        if "global_step" in pl_sd:
-            print(f"Global Step: {pl_sd['global_step']}")
-
-        sd = get_state_dict_from_checkpoint(pl_sd)
-        del pl_sd
+        sd = read_state_dict(checkpoint_file)
         model.load_state_dict(sd, strict=False)
         del sd
+        
+        if cache_enabled:
+            # cache newly loaded model
+            checkpoints_loaded[checkpoint_info] = model.state_dict().copy()
 
         if shared.cmd_opts.opt_channelslast:
             model.to(memory_format=torch.channels_last)
@@ -199,23 +218,63 @@ def load_model_weights(model, checkpoint_info, vae_file="auto"):
 
         model.first_stage_model.to(devices.dtype_vae)
 
-    else:
-        vae_name = sd_vae.get_filename(vae_file) if vae_file else None
-        vae_message = f" with {vae_name} VAE" if vae_name else ""
-        print(f"Loading weights [{sd_model_hash}]{vae_message} from cache")
-        model.load_state_dict(checkpoints_loaded[checkpoint_info])
-
-    if shared.opts.sd_checkpoint_cache > 0:
-        while len(checkpoints_loaded) > shared.opts.sd_checkpoint_cache:
+    # clean up cache if limit is reached
+    if cache_enabled:
+        while len(checkpoints_loaded) > shared.opts.sd_checkpoint_cache + 1: # we need to count the current model
             checkpoints_loaded.popitem(last=False)  # LRU
 
     model.sd_model_hash = sd_model_hash
     model.sd_model_checkpoint = checkpoint_file
     model.sd_checkpoint_info = checkpoint_info
 
+    sd_vae.delete_base_vae()
+    sd_vae.clear_loaded_vae()
+    vae_file = sd_vae.resolve_vae(checkpoint_file, vae_file=vae_file)
     sd_vae.load_vae(model, vae_file)
 
 
+def enable_midas_autodownload():
+    """
+    Gives the ldm.modules.midas.api.load_model function automatic downloading.
+
+    When the 512-depth-ema model, and other future models like it, is loaded,
+    it calls midas.api.load_model to load the associated midas depth model.
+    This function applies a wrapper to download the model to the correct
+    location automatically.
+    """
+
+    midas_path = os.path.join(models_path, 'midas')
+
+    # stable-diffusion-stability-ai hard-codes the midas model path to
+    # a location that differs from where other scripts using this model look.
+    # HACK: Overriding the path here.
+    for k, v in midas.api.ISL_PATHS.items():
+        file_name = os.path.basename(v)
+        midas.api.ISL_PATHS[k] = os.path.join(midas_path, file_name)
+
+    midas_urls = {
+        "dpt_large": "https://github.com/intel-isl/DPT/releases/download/1_0/dpt_large-midas-2f21e586.pt",
+        "dpt_hybrid": "https://github.com/intel-isl/DPT/releases/download/1_0/dpt_hybrid-midas-501f0c75.pt",
+        "midas_v21": "https://github.com/AlexeyAB/MiDaS/releases/download/midas_dpt/midas_v21-f6b98070.pt",
+        "midas_v21_small": "https://github.com/AlexeyAB/MiDaS/releases/download/midas_dpt/midas_v21_small-70d6b9c8.pt",
+    }
+
+    midas.api.load_model_inner = midas.api.load_model
+
+    def load_model_wrapper(model_type):
+        path = midas.api.ISL_PATHS[model_type]
+        if not os.path.exists(path):
+            if not os.path.exists(midas_path):
+                mkdir(midas_path)
+    
+            print(f"Downloading midas model weights for {model_type} to {path}")
+            request.urlretrieve(midas_urls[model_type], path)
+            print(f"{model_type} downloaded")
+
+        return midas.api.load_model_inner(model_type)
+
+    midas.api.load_model = load_model_wrapper
+
 def load_model(checkpoint_info=None):
     from modules import lowvram, sd_hijack
     checkpoint_info = checkpoint_info or select_checkpoint()
@@ -243,6 +302,9 @@ def load_model(checkpoint_info=None):
 
     do_inpainting_hijack()
 
+    if shared.cmd_opts.no_half:
+        sd_config.model.params.unet_config.params.use_fp16 = False
+
     sd_model = instantiate_from_config(sd_config.model)
     load_model_weights(sd_model, checkpoint_info)
 

modules/sd_samplers.py

@@ -1,4 +1,4 @@
-from collections import namedtuple
+from collections import namedtuple, deque
 import numpy as np
 from math import floor
 import torch
@@ -6,6 +6,7 @@ import tqdm
 from PIL import Image
 import inspect
 import k_diffusion.sampling
+import torchsde._brownian.brownian_interval
 import ldm.models.diffusion.ddim
 import ldm.models.diffusion.plms
 from modules import prompt_parser, devices, processing, images
@@ -18,7 +19,7 @@ from modules.script_callbacks import CFGDenoiserParams, cfg_denoiser_callback
 SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options'])
 
 samplers_k_diffusion = [
-    ('Euler a', 'sample_euler_ancestral', ['k_euler_a'], {}),
+    ('Euler a', 'sample_euler_ancestral', ['k_euler_a', 'k_euler_ancestral'], {}),
     ('Euler', 'sample_euler', ['k_euler'], {}),
     ('LMS', 'sample_lms', ['k_lms'], {}),
     ('Heun', 'sample_heun', ['k_heun'], {}),
@@ -26,6 +27,7 @@ samplers_k_diffusion = [
     ('DPM2 a', 'sample_dpm_2_ancestral', ['k_dpm_2_a'], {}),
     ('DPM++ 2S a', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a'], {}),
     ('DPM++ 2M', 'sample_dpmpp_2m', ['k_dpmpp_2m'], {}),
+    ('DPM++ SDE', 'sample_dpmpp_sde', ['k_dpmpp_sde'], {}),
     ('DPM fast', 'sample_dpm_fast', ['k_dpm_fast'], {}),
     ('DPM adaptive', 'sample_dpm_adaptive', ['k_dpm_ad'], {}),
     ('LMS Karras', 'sample_lms', ['k_lms_ka'], {'scheduler': 'karras'}),
@@ -33,6 +35,7 @@ samplers_k_diffusion = [
     ('DPM2 a Karras', 'sample_dpm_2_ancestral', ['k_dpm_2_a_ka'], {'scheduler': 'karras'}),
     ('DPM++ 2S a Karras', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a_ka'], {'scheduler': 'karras'}),
     ('DPM++ 2M Karras', 'sample_dpmpp_2m', ['k_dpmpp_2m_ka'], {'scheduler': 'karras'}),
+    ('DPM++ SDE Karras', 'sample_dpmpp_sde', ['k_dpmpp_sde_ka'], {'scheduler': 'karras'}),
 ]
 
 samplers_data_k_diffusion = [
@@ -46,16 +49,24 @@ all_samplers = [
     SamplerData('DDIM', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.ddim.DDIMSampler, model), [], {}),
     SamplerData('PLMS', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.plms.PLMSSampler, model), [], {}),
 ]
+all_samplers_map = {x.name: x for x in all_samplers}
 
 samplers = []
 samplers_for_img2img = []
+samplers_map = {}
 
 
-def create_sampler_with_index(list_of_configs, index, model):
-    config = list_of_configs[index]
+def create_sampler(name, model):
+    if name is not None:
+        config = all_samplers_map.get(name, None)
+    else:
+        config = all_samplers[0]
+
+    assert config is not None, f'bad sampler name: {name}'
+
     sampler = config.constructor(model)
     sampler.config = config
-    
+
     return sampler
 
 
@@ -68,6 +79,12 @@ def set_samplers():
     samplers = [x for x in all_samplers if x.name not in hidden]
     samplers_for_img2img = [x for x in all_samplers if x.name not in hidden_img2img]
 
+    samplers_map.clear()
+    for sampler in all_samplers:
+        samplers_map[sampler.name.lower()] = sampler.name
+        for alias in sampler.aliases:
+            samplers_map[alias.lower()] = sampler.name
+
 
 set_samplers()
 
@@ -120,7 +137,8 @@ class InterruptedException(BaseException):
 class VanillaStableDiffusionSampler:
     def __init__(self, constructor, sd_model):
         self.sampler = constructor(sd_model)
-        self.orig_p_sample_ddim = self.sampler.p_sample_ddim if hasattr(self.sampler, 'p_sample_ddim') else self.sampler.p_sample_plms
+        self.is_plms = hasattr(self.sampler, 'p_sample_plms')
+        self.orig_p_sample_ddim = self.sampler.p_sample_plms if self.is_plms else self.sampler.p_sample_ddim
         self.mask = None
         self.nmask = None
         self.init_latent = None
@@ -211,7 +229,6 @@ class VanillaStableDiffusionSampler:
         self.mask = p.mask if hasattr(p, 'mask') else None
         self.nmask = p.nmask if hasattr(p, 'nmask') else None
 
-
     def adjust_steps_if_invalid(self, p, num_steps):
         if  (self.config.name == 'DDIM' and p.ddim_discretize == 'uniform') or (self.config.name == 'PLMS'):
             valid_step = 999 / (1000 // num_steps)
@@ -220,7 +237,6 @@ class VanillaStableDiffusionSampler:
         
         return num_steps
 
-
     def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
         steps, t_enc = setup_img2img_steps(p, steps)
         steps = self.adjust_steps_if_invalid(p, steps)
@@ -253,9 +269,10 @@ class VanillaStableDiffusionSampler:
         steps = self.adjust_steps_if_invalid(p, steps or p.steps)
 
         # Wrap the conditioning models with additional image conditioning for inpainting model
+        # dummy_for_plms is needed because PLMS code checks the first item in the dict to have the right shape
         if image_conditioning is not None:
-            conditioning = {"c_concat": [image_conditioning], "c_crossattn": [conditioning]}
-            unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]}
+            conditioning = {"dummy_for_plms": np.zeros((conditioning.shape[0],)), "c_crossattn": [conditioning], "c_concat": [image_conditioning]}
+            unconditional_conditioning = {"c_crossattn": [unconditional_conditioning], "c_concat": [image_conditioning]}
 
         samples_ddim = self.launch_sampling(steps, lambda: self.sampler.sample(S=steps, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=self.eta)[0])
 
@@ -328,28 +345,55 @@ class CFGDenoiser(torch.nn.Module):
 
 
 class TorchHijack:
-    def __init__(self, kdiff_sampler):
-        self.kdiff_sampler = kdiff_sampler
+    def __init__(self, sampler_noises):
+        # Using a deque to efficiently receive the sampler_noises in the same order as the previous index-based
+        # implementation.
+        self.sampler_noises = deque(sampler_noises)
 
     def __getattr__(self, item):
         if item == 'randn_like':
-            return self.kdiff_sampler.randn_like
+            return self.randn_like
 
         if hasattr(torch, item):
             return getattr(torch, item)
 
         raise AttributeError("'{}' object has no attribute '{}'".format(type(self).__name__, item))
 
+    def randn_like(self, x):
+        if self.sampler_noises:
+            noise = self.sampler_noises.popleft()
+            if noise.shape == x.shape:
+                return noise
+
+        if x.device.type == 'mps':
+            return torch.randn_like(x, device=devices.cpu).to(x.device)
+        else:
+            return torch.randn_like(x)
+
+
+# MPS fix for randn in torchsde
+def torchsde_randn(size, dtype, device, seed):
+    if device.type == 'mps':
+        generator = torch.Generator(devices.cpu).manual_seed(int(seed))
+        return torch.randn(size, dtype=dtype, device=devices.cpu, generator=generator).to(device)
+    else:
+        generator = torch.Generator(device).manual_seed(int(seed))
+        return torch.randn(size, dtype=dtype, device=device, generator=generator)
+
+
+torchsde._brownian.brownian_interval._randn = torchsde_randn
+
 
 class KDiffusionSampler:
     def __init__(self, funcname, sd_model):
-        self.model_wrap = k_diffusion.external.CompVisDenoiser(sd_model, quantize=shared.opts.enable_quantization)
+        denoiser = k_diffusion.external.CompVisVDenoiser if sd_model.parameterization == "v" else k_diffusion.external.CompVisDenoiser
+
+        self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization)
         self.funcname = funcname
         self.func = getattr(k_diffusion.sampling, self.funcname)
         self.extra_params = sampler_extra_params.get(funcname, [])
         self.model_wrap_cfg = CFGDenoiser(self.model_wrap)
         self.sampler_noises = None
-        self.sampler_noise_index = 0
         self.stop_at = None
         self.eta = None
         self.default_eta = 1.0
@@ -382,26 +426,13 @@ class KDiffusionSampler:
     def number_of_needed_noises(self, p):
         return p.steps
 
-    def randn_like(self, x):
-        noise = self.sampler_noises[self.sampler_noise_index] if self.sampler_noises is not None and self.sampler_noise_index < len(self.sampler_noises) else None
-
-        if noise is not None and x.shape == noise.shape:
-            res = noise
-        else:
-            res = torch.randn_like(x)
-
-        self.sampler_noise_index += 1
-        return res
-
     def initialize(self, p):
         self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None
         self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None
         self.model_wrap.step = 0
-        self.sampler_noise_index = 0
         self.eta = p.eta or opts.eta_ancestral
 
-        if self.sampler_noises is not None:
-            k_diffusion.sampling.torch = TorchHijack(self)
+        k_diffusion.sampling.torch = TorchHijack(self.sampler_noises if self.sampler_noises is not None else [])
 
         extra_params_kwargs = {}
         for param_name in self.extra_params:

modules/sd_vae.py

@@ -4,6 +4,7 @@ from collections import namedtuple
 from modules import shared, devices, script_callbacks
 from modules.paths import models_path
 import glob
+from copy import deepcopy
 
 
 model_dir = "Stable-diffusion"
@@ -15,7 +16,7 @@ vae_path = os.path.abspath(os.path.join(models_path, vae_dir))
 vae_ignore_keys = {"model_ema.decay", "model_ema.num_updates"}
 
 
-default_vae_dict = {"auto": "auto", "None": "None"}
+default_vae_dict = {"auto": "auto", "None": None, None: None}
 default_vae_list = ["auto", "None"]
 
 
@@ -39,7 +40,8 @@ def get_base_vae(model):
 def store_base_vae(model):
     global base_vae, checkpoint_info
     if checkpoint_info != model.sd_checkpoint_info:
-        base_vae = model.first_stage_model.state_dict().copy()
+        assert not loaded_vae_file, "Trying to store non-base VAE!"
+        base_vae = deepcopy(model.first_stage_model.state_dict())
         checkpoint_info = model.sd_checkpoint_info
 
 
@@ -50,9 +52,11 @@ def delete_base_vae():
 
 
 def restore_base_vae(model):
-    global base_vae, checkpoint_info
+    global loaded_vae_file
     if base_vae is not None and checkpoint_info == model.sd_checkpoint_info:
-        load_vae_dict(model, base_vae)
+        print("Restoring base VAE")
+        _load_vae_dict(model, base_vae)
+        loaded_vae_file = None
     delete_base_vae()
 
 
@@ -83,47 +87,54 @@ def refresh_vae_list(vae_path=vae_path, model_path=model_path):
     return vae_list
 
 
-def resolve_vae(checkpoint_file, vae_file="auto"):
+def get_vae_from_settings(vae_file="auto"):
+    # else, we load from settings, if not set to be default
+    if vae_file == "auto" and shared.opts.sd_vae is not None:
+        # if saved VAE settings isn't recognized, fallback to auto
+        vae_file = vae_dict.get(shared.opts.sd_vae, "auto")
+        # if VAE selected but not found, fallback to auto
+        if vae_file not in default_vae_values and not os.path.isfile(vae_file):
+            vae_file = "auto"
+            print(f"Selected VAE doesn't exist: {vae_file}")
+    return vae_file
+
+
+def resolve_vae(checkpoint_file=None, vae_file="auto"):
     global first_load, vae_dict, vae_list
 
     # if vae_file argument is provided, it takes priority, but not saved
     if vae_file and vae_file not in default_vae_list:
         if not os.path.isfile(vae_file):
+            print(f"VAE provided as function argument doesn't exist: {vae_file}")
             vae_file = "auto"
-            print("VAE provided as function argument doesn't exist")
     # for the first load, if vae-path is provided, it takes priority, saved, and failure is reported
     if first_load and shared.cmd_opts.vae_path is not None:
         if os.path.isfile(shared.cmd_opts.vae_path):
             vae_file = shared.cmd_opts.vae_path
             shared.opts.data['sd_vae'] = get_filename(vae_file)
         else:
-            print("VAE provided as command line argument doesn't exist")
-    # else, we load from settings
-    if vae_file == "auto" and shared.opts.sd_vae is not None:
-        # if saved VAE settings isn't recognized, fallback to auto
-        vae_file = vae_dict.get(shared.opts.sd_vae, "auto")
-        # if VAE selected but not found, fallback to auto
-        if vae_file not in default_vae_values and not os.path.isfile(vae_file):
-            vae_file = "auto"
-            print("Selected VAE doesn't exist")
+            print(f"VAE provided as command line argument doesn't exist: {vae_file}")
+    # fallback to selector in settings, if vae selector not set to act as default fallback
+    if not shared.opts.sd_vae_as_default:
+        vae_file = get_vae_from_settings(vae_file)
     # vae-path cmd arg takes priority for auto
     if vae_file == "auto" and shared.cmd_opts.vae_path is not None:
         if os.path.isfile(shared.cmd_opts.vae_path):
             vae_file = shared.cmd_opts.vae_path
-            print("Using VAE provided as command line argument")
+            print(f"Using VAE provided as command line argument: {vae_file}")
     # if still not found, try look for ".vae.pt" beside model
     model_path = os.path.splitext(checkpoint_file)[0]
     if vae_file == "auto":
         vae_file_try = model_path + ".vae.pt"
         if os.path.isfile(vae_file_try):
             vae_file = vae_file_try
-            print("Using VAE found beside selected model")
+            print(f"Using VAE found similar to selected model: {vae_file}")
     # if still not found, try look for ".vae.ckpt" beside model
     if vae_file == "auto":
         vae_file_try = model_path + ".vae.ckpt"
         if os.path.isfile(vae_file_try):
             vae_file = vae_file_try
-            print("Using VAE found beside selected model")
+            print(f"Using VAE found similar to selected model: {vae_file}")
     # No more fallbacks for auto
     if vae_file == "auto":
         vae_file = None
@@ -139,10 +150,12 @@ def load_vae(model, vae_file=None):
     # save_settings = False
 
     if vae_file:
+        assert os.path.isfile(vae_file), f"VAE file doesn't exist: {vae_file}"
         print(f"Loading VAE weights from: {vae_file}")
+        store_base_vae(model)
         vae_ckpt = torch.load(vae_file, map_location=shared.weight_load_location)
         vae_dict_1 = {k: v for k, v in vae_ckpt["state_dict"].items() if k[0:4] != "loss" and k not in vae_ignore_keys}
-        load_vae_dict(model, vae_dict_1)
+        _load_vae_dict(model, vae_dict_1)
 
         # If vae used is not in dict, update it
         # It will be removed on refresh though
@@ -150,30 +163,22 @@ def load_vae(model, vae_file=None):
         if vae_opt not in vae_dict:
             vae_dict[vae_opt] = vae_file
             vae_list.append(vae_opt)
+    elif loaded_vae_file:
+        restore_base_vae(model)
 
     loaded_vae_file = vae_file
 
-    """
-    # Save current VAE to VAE settings, maybe? will it work?
-    if save_settings:
-        if vae_file is None:
-            vae_opt = "None"
-
-        # shared.opts.sd_vae = vae_opt
-    """
-
     first_load = False
 
 
 # don't call this from outside
-def load_vae_dict(model, vae_dict_1=None):
-    if vae_dict_1:
-        store_base_vae(model)
-        model.first_stage_model.load_state_dict(vae_dict_1)
-    else:
-        restore_base_vae()
+def _load_vae_dict(model, vae_dict_1):
+    model.first_stage_model.load_state_dict(vae_dict_1)
     model.first_stage_model.to(devices.dtype_vae)
 
+def clear_loaded_vae():
+    global loaded_vae_file
+    loaded_vae_file = None
 
 def reload_vae_weights(sd_model=None, vae_file="auto"):
     from modules import lowvram, devices, sd_hijack

modules/shared.py

@@ -3,7 +3,6 @@ import datetime
 import json
 import os
 import sys
-from collections import OrderedDict
 import time
 
 import gradio as gr
@@ -12,17 +11,18 @@ import tqdm
 import modules.artists
 import modules.interrogate
 import modules.memmon
-import modules.sd_models
 import modules.styles
 import modules.devices as devices
-from modules import sd_samplers, sd_models, localization, sd_vae
-from modules.hypernetworks import hypernetwork
+from modules import localization, sd_vae, extensions, script_loading
 from modules.paths import models_path, script_path, sd_path
 
+
+demo = None
+
 sd_model_file = os.path.join(script_path, 'model.ckpt')
 default_sd_model_file = sd_model_file
 parser = argparse.ArgumentParser()
-parser.add_argument("--config", type=str, default=os.path.join(sd_path, "configs/stable-diffusion/v1-inference.yaml"), help="path to config which constructs model",)
+parser.add_argument("--config", type=str, default=os.path.join(script_path, "v1-inference.yaml"), help="path to config which constructs model",)
 parser.add_argument("--ckpt", type=str, default=sd_model_file, help="path to checkpoint of stable diffusion model; if specified, this checkpoint will be added to the list of checkpoints and loaded",)
 parser.add_argument("--ckpt-dir", type=str, default=None, help="Path to directory with stable diffusion checkpoints")
 parser.add_argument("--gfpgan-dir", type=str, help="GFPGAN directory", default=('./src/gfpgan' if os.path.exists('./src/gfpgan') else './GFPGAN'))
@@ -50,18 +50,15 @@ parser.add_argument("--gfpgan-models-path", type=str, help="Path to directory wi
 parser.add_argument("--esrgan-models-path", type=str, help="Path to directory with ESRGAN model file(s).", default=os.path.join(models_path, 'ESRGAN'))
 parser.add_argument("--bsrgan-models-path", type=str, help="Path to directory with BSRGAN model file(s).", default=os.path.join(models_path, 'BSRGAN'))
 parser.add_argument("--realesrgan-models-path", type=str, help="Path to directory with RealESRGAN model file(s).", default=os.path.join(models_path, 'RealESRGAN'))
-parser.add_argument("--scunet-models-path", type=str, help="Path to directory with ScuNET model file(s).", default=os.path.join(models_path, 'ScuNET'))
-parser.add_argument("--swinir-models-path", type=str, help="Path to directory with SwinIR model file(s).", default=os.path.join(models_path, 'SwinIR'))
-parser.add_argument("--ldsr-models-path", type=str, help="Path to directory with LDSR model file(s).", default=os.path.join(models_path, 'LDSR'))
 parser.add_argument("--clip-models-path", type=str, help="Path to directory with CLIP model file(s).", default=None)
 parser.add_argument("--xformers", action='store_true', help="enable xformers for cross attention layers")
 parser.add_argument("--force-enable-xformers", action='store_true', help="enable xformers for cross attention layers regardless of whether the checking code thinks you can run it; do not make bug reports if this fails to work")
-parser.add_argument("--deepdanbooru", action='store_true', help="enable deepdanbooru interrogator")
+parser.add_argument("--deepdanbooru", action='store_true', help="does not do anything")
 parser.add_argument("--opt-split-attention", action='store_true', help="force-enables Doggettx's cross-attention layer optimization. By default, it's on for torch cuda.")
 parser.add_argument("--opt-split-attention-invokeai", action='store_true', help="force-enables InvokeAI's cross-attention layer optimization. By default, it's on when cuda is unavailable.")
 parser.add_argument("--opt-split-attention-v1", action='store_true', help="enable older version of split attention optimization that does not consume all the VRAM it can find")
 parser.add_argument("--disable-opt-split-attention", action='store_true', help="force-disables cross-attention layer optimization")
-parser.add_argument("--use-cpu", nargs='+',choices=['all', 'sd', 'interrogate', 'gfpgan', 'swinir', 'esrgan', 'scunet', 'codeformer'], help="use CPU as torch device for specified modules", default=[], type=str.lower)
+parser.add_argument("--use-cpu", nargs='+', help="use CPU as torch device for specified modules", default=[], type=str.lower)
 parser.add_argument("--listen", action='store_true', help="launch gradio with 0.0.0.0 as server name, allowing to respond to network requests")
 parser.add_argument("--port", type=int, help="launch gradio with given server port, you need root/admin rights for ports < 1024, defaults to 7860 if available", default=None)
 parser.add_argument("--show-negative-prompt", action='store_true', help="does not do anything", default=False)
@@ -72,6 +69,7 @@ parser.add_argument("--ui-settings-file", type=str, help="filename to use for ui
 parser.add_argument("--gradio-debug",  action='store_true', help="launch gradio with --debug option")
 parser.add_argument("--gradio-auth", type=str, help='set gradio authentication like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"', default=None)
 parser.add_argument("--gradio-img2img-tool", type=str, help='gradio image uploader tool: can be either editor for ctopping, or color-sketch for drawing', choices=["color-sketch", "editor"], default="editor")
+parser.add_argument("--gradio-inpaint-tool", type=str, choices=["sketch", "color-sketch"], default="sketch", help="gradio inpainting editor: can be either sketch to only blur/noise the input, or color-sketch to paint over it")
 parser.add_argument("--opt-channelslast", action='store_true', help="change memory type for stable diffusion to channels last")
 parser.add_argument("--styles-file", type=str, help="filename to use for styles", default=os.path.join(script_path, 'styles.csv'))
 parser.add_argument("--autolaunch", action='store_true', help="open the webui URL in the system's default browser upon launch", default=False)
@@ -81,17 +79,23 @@ parser.add_argument("--disable-console-progressbars", action='store_true', help=
 parser.add_argument("--enable-console-prompts", action='store_true', help="print prompts to console when generating with txt2img and img2img", default=False)
 parser.add_argument('--vae-path', type=str, help='Path to Variational Autoencoders model', default=None)
 parser.add_argument("--disable-safe-unpickle", action='store_true', help="disable checking pytorch models for malicious code", default=False)
-parser.add_argument("--api", action='store_true', help="use api=True to launch the api with the webui")
-parser.add_argument("--nowebui", action='store_true', help="use api=True to launch the api instead of the webui")
+parser.add_argument("--api", action='store_true', help="use api=True to launch the API together with the webui (use --nowebui instead for only the API)")
+parser.add_argument("--api-auth", type=str, help='Set authentication for API like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"', default=None)
+parser.add_argument("--nowebui", action='store_true', help="use api=True to launch the API instead of the webui")
 parser.add_argument("--ui-debug-mode", action='store_true', help="Don't load model to quickly launch UI")
 parser.add_argument("--device-id", type=str, help="Select the default CUDA device to use (export CUDA_VISIBLE_DEVICES=0,1,etc might be needed before)", default=None)
 parser.add_argument("--administrator", action='store_true', help="Administrator rights", default=False)
-parser.add_argument("--cors-allow-origins", type=str, help="Allowed CORS origins", default=None)
+parser.add_argument("--cors-allow-origins", type=str, help="Allowed CORS origin(s) in the form of a comma-separated list (no spaces)", default=None)
+parser.add_argument("--cors-allow-origins-regex", type=str, help="Allowed CORS origin(s) in the form of a single regular expression", default=None)
 parser.add_argument("--tls-keyfile", type=str, help="Partially enables TLS, requires --tls-certfile to fully function", default=None)
 parser.add_argument("--tls-certfile", type=str, help="Partially enables TLS, requires --tls-keyfile to fully function", default=None)
 parser.add_argument("--server-name", type=str, help="Sets hostname of server", default=None)
 
+script_loading.preload_extensions(extensions.extensions_dir, parser)
+script_loading.preload_extensions(extensions.extensions_builtin_dir, parser)
+
 cmd_opts = parser.parse_args()
+
 restricted_opts = {
     "samples_filename_pattern",
     "directories_filename_pattern",
@@ -104,10 +108,10 @@ restricted_opts = {
     "outdir_save",
 }
 
-cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen) and not cmd_opts.enable_insecure_extension_access
+cmd_opts.disable_extension_access = (cmd_opts.share or cmd_opts.listen or cmd_opts.server_name) and not cmd_opts.enable_insecure_extension_access
 
-devices.device, devices.device_interrogate, devices.device_gfpgan, devices.device_swinir, devices.device_esrgan, devices.device_scunet, devices.device_codeformer = \
-(devices.cpu if any(y in cmd_opts.use_cpu for y in [x, 'all']) else devices.get_optimal_device() for x in ['sd', 'interrogate', 'gfpgan', 'swinir', 'esrgan', 'scunet', 'codeformer'])
+devices.device, devices.device_interrogate, devices.device_gfpgan, devices.device_esrgan, devices.device_codeformer = \
+    (devices.cpu if any(y in cmd_opts.use_cpu for y in [x, 'all']) else devices.get_optimal_device() for x in ['sd', 'interrogate', 'gfpgan', 'esrgan', 'codeformer'])
 
 device = devices.device
 weight_load_location = None if cmd_opts.lowram else "cpu"
@@ -118,10 +122,12 @@ xformers_available = False
 config_filename = cmd_opts.ui_settings_file
 
 os.makedirs(cmd_opts.hypernetwork_dir, exist_ok=True)
-hypernetworks = hypernetwork.list_hypernetworks(cmd_opts.hypernetwork_dir)
+hypernetworks = {}
 loaded_hypernetwork = None
 
+
 def reload_hypernetworks():
+    from modules.hypernetworks import hypernetwork
     global hypernetworks
 
     hypernetworks = hypernetwork.list_hypernetworks(cmd_opts.hypernetwork_dir)
@@ -203,10 +209,11 @@ class State:
         if self.current_latent is None:
             return
 
+        import modules.sd_samplers
         if opts.show_progress_grid:
-            self.current_image = sd_samplers.samples_to_image_grid(self.current_latent)
+            self.current_image = modules.sd_samplers.samples_to_image_grid(self.current_latent)
         else:
-            self.current_image = sd_samplers.sample_to_image(self.current_latent)
+            self.current_image = modules.sd_samplers.sample_to_image(self.current_latent)
 
         self.current_image_sampling_step = self.sampling_step
 
@@ -245,6 +252,21 @@ def options_section(section_identifier, options_dict):
     return options_dict
 
 
+def list_checkpoint_tiles():
+    import modules.sd_models
+    return modules.sd_models.checkpoint_tiles()
+
+
+def refresh_checkpoints():
+    import modules.sd_models
+    return modules.sd_models.list_models()
+
+
+def list_samplers():
+    import modules.sd_samplers
+    return modules.sd_samplers.all_samplers
+
+
 hide_dirs = {"visible": not cmd_opts.hide_ui_dir_config}
 
 options_templates = {}
@@ -273,6 +295,10 @@ options_templates.update(options_section(('saving-images', "Saving images/grids"
     "use_original_name_batch": OptionInfo(False, "Use original name for output filename during batch process in extras tab"),
     "save_selected_only": OptionInfo(True, "When using 'Save' button, only save a single selected image"),
     "do_not_add_watermark": OptionInfo(False, "Do not add watermark to images"),
+
+    "temp_dir":  OptionInfo("", "Directory for temporary images; leave empty for default"),
+    "clean_temp_dir_at_start": OptionInfo(False, "Cleanup non-default temporary directory when starting webui"),
+
 }))
 
 options_templates.update(options_section(('saving-paths', "Paths for saving"), {
@@ -297,10 +323,7 @@ options_templates.update(options_section(('saving-to-dirs', "Saving to a directo
 options_templates.update(options_section(('upscaling', "Upscaling"), {
     "ESRGAN_tile": OptionInfo(192, "Tile size for ESRGAN upscalers. 0 = no tiling.", gr.Slider, {"minimum": 0, "maximum": 512, "step": 16}),
     "ESRGAN_tile_overlap": OptionInfo(8, "Tile overlap, in pixels for ESRGAN upscalers. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}),
-    "realesrgan_enabled_models": OptionInfo(["R-ESRGAN x4+", "R-ESRGAN x4+ Anime6B"], "Select which Real-ESRGAN models to show in the web UI. (Requires restart)", gr.CheckboxGroup, lambda: {"choices": realesrgan_models_names()}),
-    "SWIN_tile": OptionInfo(192, "Tile size for all SwinIR.", gr.Slider, {"minimum": 16, "maximum": 512, "step": 16}),
-    "SWIN_tile_overlap": OptionInfo(8, "Tile overlap, in pixels for SwinIR. Low values = visible seam.", gr.Slider, {"minimum": 0, "maximum": 48, "step": 1}),
-    "ldsr_steps": OptionInfo(100, "LDSR processing steps. Lower = faster", gr.Slider, {"minimum": 1, "maximum": 200, "step": 1}),
+    "realesrgan_enabled_models": OptionInfo(["R-ESRGAN 4x+", "R-ESRGAN 4x+ Anime6B"], "Select which Real-ESRGAN models to show in the web UI. (Requires restart)", gr.CheckboxGroup, lambda: {"choices": realesrgan_models_names()}),
     "upscaler_for_img2img": OptionInfo(None, "Upscaler for img2img", gr.Dropdown, lambda: {"choices": [x.name for x in sd_upscalers]}),
     "use_scale_latent_for_hires_fix": OptionInfo(False, "Upscale latent space image when doing hires. fix"),
 }))
@@ -319,6 +342,7 @@ options_templates.update(options_section(('system', "System"), {
 
 options_templates.update(options_section(('training', "Training"), {
     "unload_models_when_training": OptionInfo(False, "Move VAE and CLIP to RAM when training if possible. Saves VRAM."),
+    "pin_memory": OptionInfo(False, "Turn on pin_memory for DataLoader. Makes training slightly faster but can increase memory usage."),
     "save_optimizer_state": OptionInfo(False, "Saves Optimizer state as separate *.optim file. Training can be resumed with HN itself and matching optim file."),
     "dataset_filename_word_regex": OptionInfo("", "Filename word regex"),
     "dataset_filename_join_string": OptionInfo(" ", "Filename join string"),
@@ -328,12 +352,14 @@ options_templates.update(options_section(('training', "Training"), {
 }))
 
 options_templates.update(options_section(('sd', "Stable Diffusion"), {
-    "sd_model_checkpoint": OptionInfo(None, "Stable Diffusion checkpoint", gr.Dropdown, lambda: {"choices": modules.sd_models.checkpoint_tiles()}, refresh=sd_models.list_models),
+    "sd_model_checkpoint": OptionInfo(None, "Stable Diffusion checkpoint", gr.Dropdown, lambda: {"choices": list_checkpoint_tiles()}, refresh=refresh_checkpoints),
     "sd_checkpoint_cache": OptionInfo(0, "Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}),
-    "sd_vae": OptionInfo("auto", "SD VAE", gr.Dropdown, lambda: {"choices": list(sd_vae.vae_list)}, refresh=sd_vae.refresh_vae_list),
+    "sd_vae": OptionInfo("auto", "SD VAE", gr.Dropdown, lambda: {"choices": sd_vae.vae_list}, refresh=sd_vae.refresh_vae_list),
+    "sd_vae_as_default": OptionInfo(False, "Ignore selected VAE for stable diffusion checkpoints that have their own .vae.pt next to them"),
     "sd_hypernetwork": OptionInfo("None", "Hypernetwork", gr.Dropdown, lambda: {"choices": ["None"] + [x for x in hypernetworks.keys()]}, refresh=reload_hypernetworks),
     "sd_hypernetwork_strength": OptionInfo(1.0, "Hypernetwork strength", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.001}),
     "inpainting_mask_weight": OptionInfo(1.0, "Inpainting conditioning mask strength", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
+    "initial_noise_multiplier": OptionInfo(1.0, "Noise multiplier for img2img", gr.Slider, {"minimum": 0.5, "maximum": 1.5, "step": 0.01 }),
     "img2img_color_correction": OptionInfo(False, "Apply color correction to img2img results to match original colors."),
     "img2img_fix_steps": OptionInfo(False, "With img2img, do exactly the amount of steps the slider specifies (normally you'd do less with less denoising)."),
     "enable_quantization": OptionInfo(False, "Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply."),
@@ -342,7 +368,7 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), {
     "enable_batch_seeds": OptionInfo(True, "Make K-diffusion samplers produce same images in a batch as when making a single image"),
     "comma_padding_backtrack": OptionInfo(20, "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens", gr.Slider, {"minimum": 0, "maximum": 74, "step": 1 }),
     "filter_nsfw": OptionInfo(False, "Filter NSFW content"),
-    'CLIP_stop_at_last_layers': OptionInfo(1, "Stop At last layers of CLIP model", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}),
+    'CLIP_stop_at_last_layers': OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}),
     "random_artist_categories": OptionInfo([], "Allowed categories for random artists selection when using the Roll button", gr.CheckboxGroup, {"choices": artist_db.categories()}),
 }))
 
@@ -380,7 +406,7 @@ options_templates.update(options_section(('ui', "User interface"), {
 }))
 
 options_templates.update(options_section(('sampler-params', "Sampler parameters"), {
-    "hide_samplers": OptionInfo([], "Hide samplers in user interface (requires restart)", gr.CheckboxGroup, lambda: {"choices": [x.name for x in sd_samplers.all_samplers]}),
+    "hide_samplers": OptionInfo([], "Hide samplers in user interface (requires restart)", gr.CheckboxGroup, lambda: {"choices": [x.name for x in list_samplers()]}),
     "eta_ddim": OptionInfo(0.0, "eta (noise multiplier) for DDIM", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
     "eta_ancestral": OptionInfo(1.0, "eta (noise multiplier) for ancestral samplers", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
     "ddim_discretize": OptionInfo('uniform', "img2img DDIM discretize", gr.Radio, {"choices": ['uniform', 'quad']}),
@@ -433,6 +459,23 @@ class Options:
 
         return super(Options, self).__getattribute__(item)
 
+    def set(self, key, value):
+        """sets an option and calls its onchange callback, returning True if the option changed and False otherwise"""
+
+        oldval = self.data.get(key, None)
+        if oldval == value:
+            return False
+
+        try:
+            setattr(self, key, value)
+        except RuntimeError:
+            return False
+
+        if self.data_labels[key].onchange is not None:
+            self.data_labels[key].onchange()
+
+        return True
+
     def save(self, filename):
         assert not cmd_opts.freeze_settings, "saving settings is disabled"
 

modules/styles.py

@@ -65,17 +65,6 @@ class StyleDatabase:
     def apply_negative_styles_to_prompt(self, prompt, styles):
         return apply_styles_to_prompt(prompt, [self.styles.get(x, self.no_style).negative_prompt for x in styles])
 
-    def apply_styles(self, p: StableDiffusionProcessing) -> None:
-        if isinstance(p.prompt, list):
-            p.prompt = [self.apply_styles_to_prompt(prompt, p.styles) for prompt in p.prompt]
-        else:
-            p.prompt = self.apply_styles_to_prompt(p.prompt, p.styles)
-
-        if isinstance(p.negative_prompt, list):
-            p.negative_prompt = [self.apply_negative_styles_to_prompt(prompt, p.styles) for prompt in p.negative_prompt]
-        else:
-            p.negative_prompt = self.apply_negative_styles_to_prompt(p.negative_prompt, p.styles)
-
     def save_styles(self, path: str) -> None:
         # Write to temporary file first, so we don't nuke the file if something goes wrong
         fd, temp_path = tempfile.mkstemp(".csv")

modules/textual_inversion/autocrop.py

@@ -276,8 +276,8 @@ def poi_average(pois, settings):
         weight += poi.weight
         x += poi.x * poi.weight
         y += poi.y * poi.weight
-    avg_x = round(x / weight)
-    avg_y = round(y / weight)
+    avg_x = round(weight and x / weight)
+    avg_y = round(weight and y / weight)
 
     return PointOfInterest(avg_x, avg_y)
 
@@ -338,4 +338,4 @@ class Settings:
     self.face_points_weight = face_points_weight
     self.annotate_image = annotate_image
     self.destop_view_image = False
-    self.dnn_model_path = dnn_model_path
+    self.dnn_model_path = dnn_model_path

modules/textual_inversion/dataset.py

@@ -3,7 +3,7 @@ import numpy as np
 import PIL
 import torch
 from PIL import Image
-from torch.utils.data import Dataset
+from torch.utils.data import Dataset, DataLoader
 from torchvision import transforms
 
 import random
@@ -11,25 +11,28 @@ import tqdm
 from modules import devices, shared
 import re
 
+from ldm.modules.distributions.distributions import DiagonalGaussianDistribution
+
 re_numbers_at_start = re.compile(r"^[-\d]+\s*")
 
 
 class DatasetEntry:
-    def __init__(self, filename=None, latent=None, filename_text=None):
+    def __init__(self, filename=None, filename_text=None, latent_dist=None, latent_sample=None, cond=None, cond_text=None, pixel_values=None):
         self.filename = filename
-        self.latent = latent
         self.filename_text = filename_text
-        self.cond = None
-        self.cond_text = None
+        self.latent_dist = latent_dist
+        self.latent_sample = latent_sample
+        self.cond = cond
+        self.cond_text = cond_text
+        self.pixel_values = pixel_values
 
 
 class PersonalizedBase(Dataset):
-    def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, device=None, template_file=None, include_cond=False, batch_size=1):
+    def __init__(self, data_root, width, height, repeats, flip_p=0.5, placeholder_token="*", model=None, cond_model=None, device=None, template_file=None, include_cond=False, batch_size=1, gradient_step=1, shuffle_tags=False, tag_drop_out=0, latent_sampling_method='once'):        
         re_word = re.compile(shared.opts.dataset_filename_word_regex) if len(shared.opts.dataset_filename_word_regex) > 0 else None
-
+        
         self.placeholder_token = placeholder_token
 
-        self.batch_size = batch_size
         self.width = width
         self.height = height
         self.flip = transforms.RandomHorizontalFlip(p=flip_p)
@@ -45,11 +48,16 @@ class PersonalizedBase(Dataset):
         assert os.path.isdir(data_root), "Dataset directory doesn't exist"
         assert os.listdir(data_root), "Dataset directory is empty"
 
-        cond_model = shared.sd_model.cond_stage_model
-
         self.image_paths = [os.path.join(data_root, file_path) for file_path in os.listdir(data_root)]
+
+        
+        self.shuffle_tags = shuffle_tags
+        self.tag_drop_out = tag_drop_out
+
         print("Preparing dataset...")
         for path in tqdm.tqdm(self.image_paths):
+            if shared.state.interrupted:
+                raise Exception("inturrupted")
             try:
                 image = Image.open(path).convert('RGB').resize((self.width, self.height), PIL.Image.BICUBIC)
             except Exception:
@@ -71,53 +79,94 @@ class PersonalizedBase(Dataset):
             npimage = np.array(image).astype(np.uint8)
             npimage = (npimage / 127.5 - 1.0).astype(np.float32)
 
-            torchdata = torch.from_numpy(npimage).to(device=device, dtype=torch.float32)
-            torchdata = torch.moveaxis(torchdata, 2, 0)
-
-            init_latent = model.get_first_stage_encoding(model.encode_first_stage(torchdata.unsqueeze(dim=0))).squeeze()
-            init_latent = init_latent.to(devices.cpu)
-
-            entry = DatasetEntry(filename=path, filename_text=filename_text, latent=init_latent)
-
-            if include_cond:
+            torchdata = torch.from_numpy(npimage).permute(2, 0, 1).to(device=device, dtype=torch.float32)
+            latent_sample = None
+
+            with devices.autocast():
+                latent_dist = model.encode_first_stage(torchdata.unsqueeze(dim=0))
+
+            if latent_sampling_method == "once" or (latent_sampling_method == "deterministic" and not isinstance(latent_dist, DiagonalGaussianDistribution)):
+                latent_sample = model.get_first_stage_encoding(latent_dist).squeeze().to(devices.cpu)
+                latent_sampling_method = "once"
+                entry = DatasetEntry(filename=path, filename_text=filename_text, latent_sample=latent_sample)
+            elif latent_sampling_method == "deterministic":
+                # Works only for DiagonalGaussianDistribution
+                latent_dist.std = 0
+                latent_sample = model.get_first_stage_encoding(latent_dist).squeeze().to(devices.cpu)
+                entry = DatasetEntry(filename=path, filename_text=filename_text, latent_sample=latent_sample)
+            elif latent_sampling_method == "random":
+                entry = DatasetEntry(filename=path, filename_text=filename_text, latent_dist=latent_dist)
+
+            if not (self.tag_drop_out != 0 or self.shuffle_tags):
                 entry.cond_text = self.create_text(filename_text)
-                entry.cond = cond_model([entry.cond_text]).to(devices.cpu).squeeze(0)
 
-            self.dataset.append(entry)
-
-        assert len(self.dataset) > 0, "No images have been found in the dataset."
-        self.length = len(self.dataset) * repeats // batch_size
+            if include_cond and not (self.tag_drop_out != 0 or self.shuffle_tags):
+                with devices.autocast():
+                    entry.cond = cond_model([entry.cond_text]).to(devices.cpu).squeeze(0)
 
-        self.dataset_length = len(self.dataset)
-        self.indexes = None
-        self.shuffle()
+            self.dataset.append(entry)
+            del torchdata
+            del latent_dist
+            del latent_sample
 
-    def shuffle(self):
-        self.indexes = np.random.permutation(self.dataset_length)
+        self.length = len(self.dataset)
+        assert self.length > 0, "No images have been found in the dataset."
+        self.batch_size = min(batch_size, self.length)
+        self.gradient_step = min(gradient_step, self.length // self.batch_size)
+        self.latent_sampling_method = latent_sampling_method
 
     def create_text(self, filename_text):
         text = random.choice(self.lines)
+        tags = filename_text.split(',')
+        if self.tag_drop_out != 0:
+            tags = [t for t in tags if random.random() > self.tag_drop_out]
+        if self.shuffle_tags:
+            random.shuffle(tags)
+        text = text.replace("[filewords]", ','.join(tags))
         text = text.replace("[name]", self.placeholder_token)
-        text = text.replace("[filewords]", filename_text)
         return text
 
     def __len__(self):
         return self.length
 
     def __getitem__(self, i):
-        res = []
-
-        for j in range(self.batch_size):
-            position = i * self.batch_size + j
-            if position % len(self.indexes) == 0:
-                self.shuffle()
-
-            index = self.indexes[position % len(self.indexes)]
-            entry = self.dataset[index]
-
-            if entry.cond is None:
-                entry.cond_text = self.create_text(entry.filename_text)
-
-            res.append(entry)
-
-        return res
+        entry = self.dataset[i]
+        if self.tag_drop_out != 0 or self.shuffle_tags:
+            entry.cond_text = self.create_text(entry.filename_text)
+        if self.latent_sampling_method == "random":
+            entry.latent_sample = shared.sd_model.get_first_stage_encoding(entry.latent_dist).to(devices.cpu)
+        return entry
+
+class PersonalizedDataLoader(DataLoader):
+    def __init__(self, dataset, latent_sampling_method="once", batch_size=1, pin_memory=False):
+        super(PersonalizedDataLoader, self).__init__(dataset, shuffle=True, drop_last=True, batch_size=batch_size, pin_memory=pin_memory)
+        if latent_sampling_method == "random":
+            self.collate_fn = collate_wrapper_random
+        else:
+            self.collate_fn = collate_wrapper
+        
+
+class BatchLoader:
+    def __init__(self, data):
+        self.cond_text = [entry.cond_text for entry in data]
+        self.cond = [entry.cond for entry in data]
+        self.latent_sample = torch.stack([entry.latent_sample for entry in data]).squeeze(1)
+        #self.emb_index = [entry.emb_index for entry in data]
+        #print(self.latent_sample.device)
+
+    def pin_memory(self):
+        self.latent_sample = self.latent_sample.pin_memory()
+        return self
+
+def collate_wrapper(batch):
+    return BatchLoader(batch)
+
+class BatchLoaderRandom(BatchLoader):
+    def __init__(self, data):
+        super().__init__(data)
+
+    def pin_memory(self):
+        return self
+
+def collate_wrapper_random(batch):
+    return BatchLoaderRandom(batch)

modules/textual_inversion/preprocess.py

@@ -6,12 +6,10 @@ import sys
 import tqdm
 import time
 
-from modules import shared, images
+from modules import shared, images, deepbooru
 from modules.paths import models_path
 from modules.shared import opts, cmd_opts
 from modules.textual_inversion import autocrop
-if cmd_opts.deepdanbooru:
-    import modules.deepbooru as deepbooru
 
 
 def preprocess(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.3, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False):
@@ -20,9 +18,7 @@ def preprocess(process_src, process_dst, process_width, process_height, preproce
             shared.interrogator.load()
 
         if process_caption_deepbooru:
-            db_opts = deepbooru.create_deepbooru_opts()
-            db_opts[deepbooru.OPT_INCLUDE_RANKS] = False
-            deepbooru.create_deepbooru_process(opts.interrogate_deepbooru_score_threshold, db_opts)
+            deepbooru.model.start()
 
         preprocess_work(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_flip, process_split, process_caption, process_caption_deepbooru, split_threshold, overlap_ratio, process_focal_crop, process_focal_crop_face_weight, process_focal_crop_entropy_weight, process_focal_crop_edges_weight, process_focal_crop_debug)
 
@@ -32,9 +28,87 @@ def preprocess(process_src, process_dst, process_width, process_height, preproce
             shared.interrogator.send_blip_to_ram()
 
         if process_caption_deepbooru:
-            deepbooru.release_process()
+            deepbooru.model.stop()
 
 
+def listfiles(dirname):
+    return os.listdir(dirname)
+
+
+class PreprocessParams:
+    src = None
+    dstdir = None
+    subindex = 0
+    flip = False
+    process_caption = False
+    process_caption_deepbooru = False
+    preprocess_txt_action = None
+
+
+def save_pic_with_caption(image, index, params: PreprocessParams, existing_caption=None):
+    caption = ""
+
+    if params.process_caption:
+        caption += shared.interrogator.generate_caption(image)
+
+    if params.process_caption_deepbooru:
+        if len(caption) > 0:
+            caption += ", "
+        caption += deepbooru.model.tag_multi(image)
+
+    filename_part = params.src
+    filename_part = os.path.splitext(filename_part)[0]
+    filename_part = os.path.basename(filename_part)
+
+    basename = f"{index:05}-{params.subindex}-{filename_part}"
+    image.save(os.path.join(params.dstdir, f"{basename}.png"))
+
+    if params.preprocess_txt_action == 'prepend' and existing_caption:
+        caption = existing_caption + ' ' + caption
+    elif params.preprocess_txt_action == 'append' and existing_caption:
+        caption = caption + ' ' + existing_caption
+    elif params.preprocess_txt_action == 'copy' and existing_caption:
+        caption = existing_caption
+
+    caption = caption.strip()
+
+    if len(caption) > 0:
+        with open(os.path.join(params.dstdir, f"{basename}.txt"), "w", encoding="utf8") as file:
+            file.write(caption)
+
+    params.subindex += 1
+
+
+def save_pic(image, index, params, existing_caption=None):
+    save_pic_with_caption(image, index, params, existing_caption=existing_caption)
+
+    if params.flip:
+        save_pic_with_caption(ImageOps.mirror(image), index, params, existing_caption=existing_caption)
+
+
+def split_pic(image, inverse_xy, width, height, overlap_ratio):
+    if inverse_xy:
+        from_w, from_h = image.height, image.width
+        to_w, to_h = height, width
+    else:
+        from_w, from_h = image.width, image.height
+        to_w, to_h = width, height
+    h = from_h * to_w // from_w
+    if inverse_xy:
+        image = image.resize((h, to_w))
+    else:
+        image = image.resize((to_w, h))
+
+    split_count = math.ceil((h - to_h * overlap_ratio) / (to_h * (1.0 - overlap_ratio)))
+    y_step = (h - to_h) / (split_count - 1)
+    for i in range(split_count):
+        y = int(y_step * i)
+        if inverse_xy:
+            splitted = image.crop((y, 0, y + to_h, to_w))
+        else:
+            splitted = image.crop((0, y, to_w, y + to_h))
+        yield splitted
+
 
 def preprocess_work(process_src, process_dst, process_width, process_height, preprocess_txt_action, process_flip, process_split, process_caption, process_caption_deepbooru=False, split_threshold=0.5, overlap_ratio=0.2, process_focal_crop=False, process_focal_crop_face_weight=0.9, process_focal_crop_entropy_weight=0.3, process_focal_crop_edges_weight=0.5, process_focal_crop_debug=False):
     width = process_width
@@ -48,82 +122,28 @@ def preprocess_work(process_src, process_dst, process_width, process_height, pre
 
     os.makedirs(dst, exist_ok=True)
 
-    files = os.listdir(src)
+    files = listfiles(src)
 
     shared.state.textinfo = "Preprocessing..."
     shared.state.job_count = len(files)
 
-    def save_pic_with_caption(image, index, existing_caption=None):
-        caption = ""
-
-        if process_caption:
-            caption += shared.interrogator.generate_caption(image)
-
-        if process_caption_deepbooru:
-            if len(caption) > 0:
-                caption += ", "
-            caption += deepbooru.get_tags_from_process(image)
-
-        filename_part = filename
-        filename_part = os.path.splitext(filename_part)[0]
-        filename_part = os.path.basename(filename_part)
-
-        basename = f"{index:05}-{subindex[0]}-{filename_part}"
-        image.save(os.path.join(dst, f"{basename}.png"))
-
-        if preprocess_txt_action == 'prepend' and existing_caption:
-            caption = existing_caption + ' ' + caption
-        elif preprocess_txt_action == 'append' and existing_caption:
-            caption = caption + ' ' + existing_caption
-        elif preprocess_txt_action == 'copy' and existing_caption:
-            caption = existing_caption
-
-        caption = caption.strip()
-        
-        if len(caption) > 0:
-            with open(os.path.join(dst, f"{basename}.txt"), "w", encoding="utf8") as file:
-                file.write(caption)
-
-        subindex[0] += 1
-
-    def save_pic(image, index, existing_caption=None):
-        save_pic_with_caption(image, index, existing_caption=existing_caption)
-
-        if process_flip:
-            save_pic_with_caption(ImageOps.mirror(image), index, existing_caption=existing_caption)
-
-    def split_pic(image, inverse_xy):
-        if inverse_xy:
-            from_w, from_h = image.height, image.width
-            to_w, to_h = height, width
-        else:
-            from_w, from_h = image.width, image.height
-            to_w, to_h = width, height
-        h = from_h * to_w // from_w
-        if inverse_xy:
-            image = image.resize((h, to_w))
-        else:
-            image = image.resize((to_w, h))
-
-        split_count = math.ceil((h - to_h * overlap_ratio) / (to_h * (1.0 - overlap_ratio)))
-        y_step = (h - to_h) / (split_count - 1)
-        for i in range(split_count):
-            y = int(y_step * i)
-            if inverse_xy:
-                splitted = image.crop((y, 0, y + to_h, to_w))
-            else:
-                splitted = image.crop((0, y, to_w, y + to_h))
-            yield splitted
-
+    params = PreprocessParams()
+    params.dstdir = dst
+    params.flip = process_flip
+    params.process_caption = process_caption
+    params.process_caption_deepbooru = process_caption_deepbooru
+    params.preprocess_txt_action = preprocess_txt_action
 
     for index, imagefile in enumerate(tqdm.tqdm(files)):
-        subindex = [0]
+        params.subindex = 0
         filename = os.path.join(src, imagefile)
         try:
             img = Image.open(filename).convert("RGB")
         except Exception:
             continue
 
+        params.src = filename
+
         existing_caption = None
         existing_caption_filename = os.path.splitext(filename)[0] + '.txt'
         if os.path.exists(existing_caption_filename):
@@ -143,8 +163,8 @@ def preprocess_work(process_src, process_dst, process_width, process_height, pre
         process_default_resize = True
 
         if process_split and ratio < 1.0 and ratio <= split_threshold:
-            for splitted in split_pic(img, inverse_xy):
-                save_pic(splitted, index, existing_caption=existing_caption)
+            for splitted in split_pic(img, inverse_xy, width, height, overlap_ratio):
+                save_pic(splitted, index, params, existing_caption=existing_caption)
             process_default_resize = False
 
         if process_focal_crop and img.height != img.width:
@@ -165,11 +185,11 @@ def preprocess_work(process_src, process_dst, process_width, process_height, pre
                 dnn_model_path = dnn_model_path,
             )
             for focal in autocrop.crop_image(img, autocrop_settings):
-                save_pic(focal, index, existing_caption=existing_caption)
+                save_pic(focal, index, params, existing_caption=existing_caption)
             process_default_resize = False
 
         if process_default_resize:
             img = images.resize_image(1, img, width, height)
-            save_pic(img, index, existing_caption=existing_caption)
+            save_pic(img, index, params, existing_caption=existing_caption)
 
-        shared.state.nextjob()
+        shared.state.nextjob()

modules/textual_inversion/textual_inversion.py

@@ -10,7 +10,7 @@ import csv
 
 from PIL import Image, PngImagePlugin
 
-from modules import shared, devices, sd_hijack, processing, sd_models, images
+from modules import shared, devices, sd_hijack, processing, sd_models, images, sd_samplers
 import modules.textual_inversion.dataset
 from modules.textual_inversion.learn_schedule import LearnRateScheduler
 
@@ -64,7 +64,8 @@ class EmbeddingDatabase:
 
         self.word_embeddings[embedding.name] = embedding
 
-        ids = model.cond_stage_model.tokenizer([embedding.name], add_special_tokens=False)['input_ids'][0]
+        # TODO changing between clip and open clip changes tokenization, which will cause embeddings to stop working
+        ids = model.cond_stage_model.tokenize([embedding.name])[0]
 
         first_id = ids[0]
         if first_id not in self.ids_lookup:
@@ -155,13 +156,11 @@ class EmbeddingDatabase:
 
 def create_embedding(name, num_vectors_per_token, overwrite_old, init_text='*'):
     cond_model = shared.sd_model.cond_stage_model
-    embedding_layer = cond_model.wrapped.transformer.text_model.embeddings
 
     with devices.autocast():
         cond_model([""])  # will send cond model to GPU if lowvram/medvram is active
 
-    ids = cond_model.tokenizer(init_text, max_length=num_vectors_per_token, return_tensors="pt", add_special_tokens=False)["input_ids"]
-    embedded = embedding_layer.token_embedding.wrapped(ids.to(devices.device)).squeeze(0)
+    embedded = cond_model.encode_embedding_init_text(init_text, num_vectors_per_token)
     vec = torch.zeros((num_vectors_per_token, embedded.shape[1]), device=devices.device)
 
     for i in range(num_vectors_per_token):
@@ -184,7 +183,7 @@ def write_loss(log_directory, filename, step, epoch_len, values):
     if shared.opts.training_write_csv_every == 0:
         return
 
-    if (step + 1) % shared.opts.training_write_csv_every != 0:
+    if step % shared.opts.training_write_csv_every != 0:
         return
     write_csv_header = False if os.path.exists(os.path.join(log_directory, filename)) else True
 
@@ -194,21 +193,23 @@ def write_loss(log_directory, filename, step, epoch_len, values):
         if write_csv_header:
             csv_writer.writeheader()
 
-        epoch = step // epoch_len
-        epoch_step = step % epoch_len 
+        epoch = (step - 1) // epoch_len
+        epoch_step = (step - 1) % epoch_len 
 
         csv_writer.writerow({
-            "step": step + 1,
+            "step": step,
             "epoch": epoch,
-            "epoch_step": epoch_step + 1,
+            "epoch_step": epoch_step,
             **values,
         })
 
-def validate_train_inputs(model_name, learn_rate, batch_size, data_root, template_file, steps, save_model_every, create_image_every, log_directory, name="embedding"):
+def validate_train_inputs(model_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_model_every, create_image_every, log_directory, name="embedding"):
     assert model_name, f"{name} not selected"
     assert learn_rate, "Learning rate is empty or 0"
     assert isinstance(batch_size, int), "Batch size must be integer"
     assert batch_size > 0, "Batch size must be positive"
+    assert isinstance(gradient_step, int), "Gradient accumulation step must be integer"
+    assert gradient_step > 0, "Gradient accumulation step must be positive"
     assert data_root, "Dataset directory is empty"
     assert os.path.isdir(data_root), "Dataset directory doesn't exist"
     assert os.listdir(data_root), "Dataset directory is empty"
@@ -224,10 +225,10 @@ def validate_train_inputs(model_name, learn_rate, batch_size, data_root, templat
     if save_model_every or create_image_every:
         assert log_directory, "Log directory is empty"
 
-def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_directory, training_width, training_height, steps, create_image_every, save_embedding_every, template_file, save_image_with_stored_embedding, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
+def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_root, log_directory, training_width, training_height, steps, shuffle_tags, tag_drop_out, latent_sampling_method, create_image_every, save_embedding_every, template_file, save_image_with_stored_embedding, preview_from_txt2img, preview_prompt, preview_negative_prompt, preview_steps, preview_sampler_index, preview_cfg_scale, preview_seed, preview_width, preview_height):
     save_embedding_every = save_embedding_every or 0
     create_image_every = create_image_every or 0
-    validate_train_inputs(embedding_name, learn_rate, batch_size, data_root, template_file, steps, save_embedding_every, create_image_every, log_directory, name="embedding")
+    validate_train_inputs(embedding_name, learn_rate, batch_size, gradient_step, data_root, template_file, steps, save_embedding_every, create_image_every, log_directory, name="embedding")
 
     shared.state.textinfo = "Initializing textual inversion training..."
     shared.state.job_count = steps
@@ -255,161 +256,203 @@ def train_embedding(embedding_name, learn_rate, batch_size, data_root, log_direc
     else:
         images_embeds_dir = None
 
-    cond_model = shared.sd_model.cond_stage_model
-
     hijack = sd_hijack.model_hijack
 
     embedding = hijack.embedding_db.word_embeddings[embedding_name]
     checkpoint = sd_models.select_checkpoint()
 
-    ititial_step = embedding.step or 0
-    if ititial_step >= steps:
+    initial_step = embedding.step or 0
+    if initial_step >= steps:
         shared.state.textinfo = f"Model has already been trained beyond specified max steps"
         return embedding, filename
+    scheduler = LearnRateScheduler(learn_rate, steps, initial_step)
 
-    scheduler = LearnRateScheduler(learn_rate, steps, ititial_step)
-
-    # dataset loading may take a while, so input validations and early returns should be done before this
+   # dataset loading may take a while, so input validations and early returns should be done before this
     shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..."
-    with torch.autocast("cuda"):
-        ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=embedding_name, model=shared.sd_model, device=devices.device, template_file=template_file, batch_size=batch_size)
+    old_parallel_processing_allowed = shared.parallel_processing_allowed
+    
+    pin_memory = shared.opts.pin_memory
+    
+    ds = modules.textual_inversion.dataset.PersonalizedBase(data_root=data_root, width=training_width, height=training_height, repeats=shared.opts.training_image_repeats_per_epoch, placeholder_token=embedding_name, model=shared.sd_model, cond_model=shared.sd_model.cond_stage_model, device=devices.device, template_file=template_file, batch_size=batch_size, gradient_step=gradient_step, shuffle_tags=shuffle_tags, tag_drop_out=tag_drop_out, latent_sampling_method=latent_sampling_method)
+
+    latent_sampling_method = ds.latent_sampling_method
+
+    dl = modules.textual_inversion.dataset.PersonalizedDataLoader(ds, latent_sampling_method=latent_sampling_method, batch_size=ds.batch_size, pin_memory=pin_memory)
+
     if unload:
+        shared.parallel_processing_allowed = False
         shared.sd_model.first_stage_model.to(devices.cpu)
 
     embedding.vec.requires_grad = True
-    optimizer = torch.optim.AdamW([embedding.vec], lr=scheduler.learn_rate)
-
-    losses = torch.zeros((32,))
-
+    optimizer = torch.optim.AdamW([embedding.vec], lr=scheduler.learn_rate, weight_decay=0.0)
+    scaler = torch.cuda.amp.GradScaler()
+
+    batch_size = ds.batch_size
+    gradient_step = ds.gradient_step
+    # n steps = batch_size * gradient_step * n image processed
+    steps_per_epoch = len(ds) // batch_size // gradient_step
+    max_steps_per_epoch = len(ds) // batch_size - (len(ds) // batch_size) % gradient_step
+    loss_step = 0
+    _loss_step = 0 #internal
+
+    
     last_saved_file = "<none>"
     last_saved_image = "<none>"
     forced_filename = "<none>"
     embedding_yet_to_be_embedded = False
-
-    pbar = tqdm.tqdm(enumerate(ds), total=steps-ititial_step)
-    for i, entries in pbar:
-        embedding.step = i + ititial_step
-
-        scheduler.apply(optimizer, embedding.step)
-        if scheduler.finished:
-            break
-
-        if shared.state.interrupted:
-            break
-
-        with torch.autocast("cuda"):
-            c = cond_model([entry.cond_text for entry in entries])
-            x = torch.stack([entry.latent for entry in entries]).to(devices.device)
-            loss = shared.sd_model(x, c)[0]
-            del x
-
-            losses[embedding.step % losses.shape[0]] = loss.item()
-
-            optimizer.zero_grad()
-            loss.backward()
-            optimizer.step()
-
-        steps_done = embedding.step + 1
-
-        epoch_num = embedding.step // len(ds)
-        epoch_step = embedding.step % len(ds)
-
-        pbar.set_description(f"[Epoch {epoch_num}: {epoch_step+1}/{len(ds)}]loss: {losses.mean():.7f}")
-
-        if embedding_dir is not None and steps_done % save_embedding_every == 0:
-            # Before saving, change name to match current checkpoint.
-            embedding_name_every = f'{embedding_name}-{steps_done}'
-            last_saved_file = os.path.join(embedding_dir, f'{embedding_name_every}.pt')
-            save_embedding(embedding, checkpoint, embedding_name_every, last_saved_file, remove_cached_checksum=True)
-            embedding_yet_to_be_embedded = True
-
-        write_loss(log_directory, "textual_inversion_loss.csv", embedding.step, len(ds), {
-            "loss": f"{losses.mean():.7f}",
-            "learn_rate": scheduler.learn_rate
-        })
-
-        if images_dir is not None and steps_done % create_image_every == 0:
-            forced_filename = f'{embedding_name}-{steps_done}'
-            last_saved_image = os.path.join(images_dir, forced_filename)
-
-            shared.sd_model.first_stage_model.to(devices.device)
-
-            p = processing.StableDiffusionProcessingTxt2Img(
-                sd_model=shared.sd_model,
-                do_not_save_grid=True,
-                do_not_save_samples=True,
-                do_not_reload_embeddings=True,
-            )
-
-            if preview_from_txt2img:
-                p.prompt = preview_prompt
-                p.negative_prompt = preview_negative_prompt
-                p.steps = preview_steps
-                p.sampler_index = preview_sampler_index
-                p.cfg_scale = preview_cfg_scale
-                p.seed = preview_seed
-                p.width = preview_width
-                p.height = preview_height
-            else:
-                p.prompt = entries[0].cond_text
-                p.steps = 20
-                p.width = training_width
-                p.height = training_height
-
-            preview_text = p.prompt
-
-            processed = processing.process_images(p)
-            image = processed.images[0]
-
-            if unload:
-                shared.sd_model.first_stage_model.to(devices.cpu)
-
-            shared.state.current_image = image
-
-            if save_image_with_stored_embedding and os.path.exists(last_saved_file) and embedding_yet_to_be_embedded:
-
-                last_saved_image_chunks = os.path.join(images_embeds_dir, f'{embedding_name}-{steps_done}.png')
-
-                info = PngImagePlugin.PngInfo()
-                data = torch.load(last_saved_file)
-                info.add_text("sd-ti-embedding", embedding_to_b64(data))
-
-                title = "<{}>".format(data.get('name', '???'))
-
-                try:
-                    vectorSize = list(data['string_to_param'].values())[0].shape[0]
-                except Exception as e:
-                    vectorSize = '?'
-
-                checkpoint = sd_models.select_checkpoint()
-                footer_left = checkpoint.model_name
-                footer_mid = '[{}]'.format(checkpoint.hash)
-                footer_right = '{}v {}s'.format(vectorSize, steps_done)
-
-                captioned_image = caption_image_overlay(image, title, footer_left, footer_mid, footer_right)
-                captioned_image = insert_image_data_embed(captioned_image, data)
-
-                captioned_image.save(last_saved_image_chunks, "PNG", pnginfo=info)
-                embedding_yet_to_be_embedded = False
-
-            last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
-            last_saved_image += f", prompt: {preview_text}"
-
-        shared.state.job_no = embedding.step
-
-        shared.state.textinfo = f"""
+    
+    pbar = tqdm.tqdm(total=steps - initial_step)
+    try:
+        for i in range((steps-initial_step) * gradient_step):
+            if scheduler.finished:
+                break
+            if shared.state.interrupted:
+                break
+            for j, batch in enumerate(dl):
+                # works as a drop_last=True for gradient accumulation
+                if j == max_steps_per_epoch:
+                    break
+                scheduler.apply(optimizer, embedding.step)
+                if scheduler.finished:
+                    break
+                if shared.state.interrupted:
+                    break
+
+                with devices.autocast():
+                    # c = stack_conds(batch.cond).to(devices.device)
+                    # mask = torch.tensor(batch.emb_index).to(devices.device, non_blocking=pin_memory)
+                    # print(mask)
+                    # c[:, 1:1+embedding.vec.shape[0]] = embedding.vec.to(devices.device, non_blocking=pin_memory)
+                    x = batch.latent_sample.to(devices.device, non_blocking=pin_memory)
+                    c = shared.sd_model.cond_stage_model(batch.cond_text)
+                    loss = shared.sd_model(x, c)[0] / gradient_step
+                    del x
+                    
+                    _loss_step += loss.item()
+                scaler.scale(loss).backward()
+                
+                # go back until we reach gradient accumulation steps
+                if (j + 1) % gradient_step != 0:
+                    continue
+                scaler.step(optimizer)
+                scaler.update()
+                embedding.step += 1
+                pbar.update()
+                optimizer.zero_grad(set_to_none=True)
+                loss_step = _loss_step
+                _loss_step = 0
+
+                steps_done = embedding.step + 1
+
+                epoch_num = embedding.step // steps_per_epoch
+                epoch_step = embedding.step % steps_per_epoch
+
+                pbar.set_description(f"[Epoch {epoch_num}: {epoch_step+1}/{steps_per_epoch}]loss: {loss_step:.7f}")
+                if embedding_dir is not None and steps_done % save_embedding_every == 0:
+                    # Before saving, change name to match current checkpoint.
+                    embedding_name_every = f'{embedding_name}-{steps_done}'
+                    last_saved_file = os.path.join(embedding_dir, f'{embedding_name_every}.pt')
+                    #if shared.opts.save_optimizer_state:
+                        #embedding.optimizer_state_dict = optimizer.state_dict()
+                    save_embedding(embedding, checkpoint, embedding_name_every, last_saved_file, remove_cached_checksum=True)
+                    embedding_yet_to_be_embedded = True
+
+                write_loss(log_directory, "textual_inversion_loss.csv", embedding.step, steps_per_epoch, {
+                    "loss": f"{loss_step:.7f}",
+                    "learn_rate": scheduler.learn_rate
+                })
+
+                if images_dir is not None and steps_done % create_image_every == 0:
+                    forced_filename = f'{embedding_name}-{steps_done}'
+                    last_saved_image = os.path.join(images_dir, forced_filename)
+
+                    shared.sd_model.first_stage_model.to(devices.device)
+
+                    p = processing.StableDiffusionProcessingTxt2Img(
+                        sd_model=shared.sd_model,
+                        do_not_save_grid=True,
+                        do_not_save_samples=True,
+                        do_not_reload_embeddings=True,
+                    )
+
+                    if preview_from_txt2img:
+                        p.prompt = preview_prompt
+                        p.negative_prompt = preview_negative_prompt
+                        p.steps = preview_steps
+                        p.sampler_name = sd_samplers.samplers[preview_sampler_index].name
+                        p.cfg_scale = preview_cfg_scale
+                        p.seed = preview_seed
+                        p.width = preview_width
+                        p.height = preview_height
+                    else:
+                        p.prompt = batch.cond_text[0]
+                        p.steps = 20
+                        p.width = training_width
+                        p.height = training_height
+
+                    preview_text = p.prompt
+
+                    processed = processing.process_images(p)
+                    image = processed.images[0] if len(processed.images) > 0 else None
+
+                    if unload:
+                        shared.sd_model.first_stage_model.to(devices.cpu)
+
+                    if image is not None:
+                        shared.state.current_image = image
+                        last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
+                        last_saved_image += f", prompt: {preview_text}"
+
+                    if save_image_with_stored_embedding and os.path.exists(last_saved_file) and embedding_yet_to_be_embedded:
+
+                        last_saved_image_chunks = os.path.join(images_embeds_dir, f'{embedding_name}-{steps_done}.png')
+
+                        info = PngImagePlugin.PngInfo()
+                        data = torch.load(last_saved_file)
+                        info.add_text("sd-ti-embedding", embedding_to_b64(data))
+
+                        title = "<{}>".format(data.get('name', '???'))
+
+                        try:
+                            vectorSize = list(data['string_to_param'].values())[0].shape[0]
+                        except Exception as e:
+                            vectorSize = '?'
+
+                        checkpoint = sd_models.select_checkpoint()
+                        footer_left = checkpoint.model_name
+                        footer_mid = '[{}]'.format(checkpoint.hash)
+                        footer_right = '{}v {}s'.format(vectorSize, steps_done)
+
+                        captioned_image = caption_image_overlay(image, title, footer_left, footer_mid, footer_right)
+                        captioned_image = insert_image_data_embed(captioned_image, data)
+
+                        captioned_image.save(last_saved_image_chunks, "PNG", pnginfo=info)
+                        embedding_yet_to_be_embedded = False
+
+                    last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
+                    last_saved_image += f", prompt: {preview_text}"
+
+                shared.state.job_no = embedding.step
+
+                shared.state.textinfo = f"""
 <p>
-Loss: {losses.mean():.7f}<br/>
-Step: {embedding.step}<br/>
-Last prompt: {html.escape(entries[0].cond_text)}<br/>
+Loss: {loss_step:.7f}<br/>
+Step: {steps_done}<br/>
+Last prompt: {html.escape(batch.cond_text[0])}<br/>
 Last saved embedding: {html.escape(last_saved_file)}<br/>
 Last saved image: {html.escape(last_saved_image)}<br/>
 </p>
 """
-
-    filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
-    save_embedding(embedding, checkpoint, embedding_name, filename, remove_cached_checksum=True)
-    shared.sd_model.first_stage_model.to(devices.device)
+        filename = os.path.join(shared.cmd_opts.embeddings_dir, f'{embedding_name}.pt')
+        save_embedding(embedding, checkpoint, embedding_name, filename, remove_cached_checksum=True)
+    except Exception:
+        print(traceback.format_exc(), file=sys.stderr)
+        pass
+    finally:
+        pbar.leave = False
+        pbar.close()
+        shared.sd_model.first_stage_model.to(devices.device)
+        shared.parallel_processing_allowed = old_parallel_processing_allowed
 
     return embedding, filename
 

modules/textual_inversion/ui.py

@@ -18,7 +18,7 @@ def create_embedding(name, initialization_text, nvpt, overwrite_old):
 def preprocess(*args):
     modules.textual_inversion.preprocess.preprocess(*args)
 
-    return "Preprocessing finished.", ""
+    return f"Preprocessing {'interrupted' if shared.state.interrupted else 'finished'}.", ""
 
 
 def train_embedding(*args):

modules/txt2img.py

@@ -1,4 +1,5 @@
 import modules.scripts
+from modules import sd_samplers
 from modules.processing import StableDiffusionProcessing, Processed, StableDiffusionProcessingTxt2Img, \
     StableDiffusionProcessingImg2Img, process_images
 from modules.shared import opts, cmd_opts
@@ -21,7 +22,7 @@ def txt2img(prompt: str, negative_prompt: str, prompt_style: str, prompt_style2:
         seed_resize_from_h=seed_resize_from_h,
         seed_resize_from_w=seed_resize_from_w,
         seed_enable_extras=seed_enable_extras,
-        sampler_index=sampler_index,
+        sampler_name=sd_samplers.samplers[sampler_index].name,
         batch_size=batch_size,
         n_iter=n_iter,
         steps=steps,

modules/ui.py

@@ -17,21 +17,17 @@ import gradio.routes
 import gradio.utils
 import numpy as np
 from PIL import Image, PngImagePlugin
+from modules.call_queue import wrap_gradio_gpu_call, wrap_queued_call, wrap_gradio_call
 
-
-from modules import sd_hijack, sd_models, localization, script_callbacks, ui_extensions
+from modules import sd_hijack, sd_models, localization, script_callbacks, ui_extensions, deepbooru
 from modules.paths import script_path
 
 from modules.shared import opts, cmd_opts, restricted_opts
 
-if cmd_opts.deepdanbooru:
-    from modules.deepbooru import get_deepbooru_tags
-
 import modules.codeformer_model
 import modules.generation_parameters_copypaste as parameters_copypaste
 import modules.gfpgan_model
 import modules.hypernetworks.ui
-import modules.ldsr_model
 import modules.scripts
 import modules.shared as shared
 import modules.styles
@@ -69,8 +65,11 @@ sample_img2img = sample_img2img if os.path.exists(sample_img2img) else None
 css_hide_progressbar = """
 .wrap .m-12 svg { display:none!important; }
 .wrap .m-12::before { content:"Loading..." }
+.wrap .z-20 svg { display:none!important; }
+.wrap .z-20::before { content:"Loading..." }
 .progress-bar { display:none!important; }
 .meta-text { display:none!important; }
+.meta-text-center { display:none!important; }
 """
 
 # Using constants for these since the variation selector isn't visible.
@@ -143,7 +142,7 @@ def save_files(js_data, images, do_make_zip, index):
                 filenames.append(os.path.basename(txt_fullfn))
                 fullfns.append(txt_fullfn)
 
-        writer.writerow([data["prompt"], data["seed"], data["width"], data["height"], data["sampler"], data["cfg_scale"], data["steps"], filenames[0], data["negative_prompt"]])
+        writer.writerow([data["prompt"], data["seed"], data["width"], data["height"], data["sampler_name"], data["cfg_scale"], data["steps"], filenames[0], data["negative_prompt"]])
 
     # Make Zip
     if do_make_zip:
@@ -158,84 +157,7 @@ def save_files(js_data, images, do_make_zip, index):
 
     return gr.File.update(value=fullfns, visible=True), '', '', plaintext_to_html(f"Saved: {filenames[0]}")
 
-def save_pil_to_file(pil_image, dir=None):
-    use_metadata = False
-    metadata = PngImagePlugin.PngInfo()
-    for key, value in pil_image.info.items():
-        if isinstance(key, str) and isinstance(value, str):
-            metadata.add_text(key, value)
-            use_metadata = True
-
-    file_obj = tempfile.NamedTemporaryFile(delete=False, suffix=".png", dir=dir)
-    pil_image.save(file_obj, pnginfo=(metadata if use_metadata else None))
-    return file_obj
-
-
-# override save to file function so that it also writes PNG info
-gr.processing_utils.save_pil_to_file = save_pil_to_file
-
-
-def wrap_gradio_call(func, extra_outputs=None, add_stats=False):
-    def f(*args, extra_outputs_array=extra_outputs, **kwargs):
-        run_memmon = opts.memmon_poll_rate > 0 and not shared.mem_mon.disabled and add_stats
-        if run_memmon:
-            shared.mem_mon.monitor()
-        t = time.perf_counter()
-
-        try:
-            res = list(func(*args, **kwargs))
-        except Exception as e:
-            # When printing out our debug argument list, do not print out more than a MB of text
-            max_debug_str_len = 131072 # (1024*1024)/8
-
-            print("Error completing request", file=sys.stderr)
-            argStr = f"Arguments: {str(args)} {str(kwargs)}"
-            print(argStr[:max_debug_str_len], file=sys.stderr)
-            if len(argStr) > max_debug_str_len:
-                print(f"(Argument list truncated at {max_debug_str_len}/{len(argStr)} characters)", file=sys.stderr)
-
-            print(traceback.format_exc(), file=sys.stderr)
-
-            shared.state.job = ""
-            shared.state.job_count = 0
-
-            if extra_outputs_array is None:
-                extra_outputs_array = [None, '']
-
-            res = extra_outputs_array + [f"<div class='error'>{plaintext_to_html(type(e).__name__+': '+str(e))}</div>"]
-
-        shared.state.skipped = False
-        shared.state.interrupted = False
-        shared.state.job_count = 0
-
-        if not add_stats:
-            return tuple(res)
-
-        elapsed = time.perf_counter() - t
-        elapsed_m = int(elapsed // 60)
-        elapsed_s = elapsed % 60
-        elapsed_text = f"{elapsed_s:.2f}s"
-        if elapsed_m > 0:
-            elapsed_text = f"{elapsed_m}m "+elapsed_text
-
-        if run_memmon:
-            mem_stats = {k: -(v//-(1024*1024)) for k, v in shared.mem_mon.stop().items()}
-            active_peak = mem_stats['active_peak']
-            reserved_peak = mem_stats['reserved_peak']
-            sys_peak = mem_stats['system_peak']
-            sys_total = mem_stats['total']
-            sys_pct = round(sys_peak/max(sys_total, 1) * 100, 2)
-
-            vram_html = f"<p class='vram'>Torch active/reserved: {active_peak}/{reserved_peak} MiB, <wbr>Sys VRAM: {sys_peak}/{sys_total} MiB ({sys_pct}%)</p>"
-        else:
-            vram_html = ''
-
-        # last item is always HTML
-        res[-1] += f"<div class='performance'><p class='time'>Time taken: <wbr>{elapsed_text}</p>{vram_html}</div>"
 
-        return tuple(res)
-
-    return f
 
 
 def calc_time_left(progress, threshold, label, force_display):
@@ -350,7 +272,7 @@ def interrogate(image):
 
 
 def interrogate_deepbooru(image):
-    prompt = get_deepbooru_tags(image)
+    prompt = deepbooru.model.tag(image)
     return gr_show(True) if prompt is None else prompt
 
 
@@ -381,8 +303,8 @@ def create_seed_inputs():
 
     with gr.Row(visible=False) as seed_extra_row_2:
         seed_extras.append(seed_extra_row_2)
-        seed_resize_from_w = gr.Slider(minimum=0, maximum=2048, step=64, label="Resize seed from width", value=0)
-        seed_resize_from_h = gr.Slider(minimum=0, maximum=2048, step=64, label="Resize seed from height", value=0)
+        seed_resize_from_w = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize seed from width", value=0)
+        seed_resize_from_h = gr.Slider(minimum=0, maximum=2048, step=8, label="Resize seed from height", value=0)
 
     random_seed.click(fn=lambda: -1, show_progress=False, inputs=[], outputs=[seed])
     random_subseed.click(fn=lambda: -1, show_progress=False, inputs=[], outputs=[subseed])
@@ -496,9 +418,7 @@ def create_toprow(is_img2img):
         if is_img2img:
             with gr.Column(scale=1, elem_id="interrogate_col"):
                 button_interrogate = gr.Button('Interrogate\nCLIP', elem_id="interrogate")
-
-                if cmd_opts.deepdanbooru:
-                    button_deepbooru = gr.Button('Interrogate\nDeepBooru', elem_id="deepbooru")
+                button_deepbooru = gr.Button('Interrogate\nDeepBooru', elem_id="deepbooru")
 
         with gr.Column(scale=1):
             with gr.Row():
@@ -584,6 +504,19 @@ def apply_setting(key, value):
     return value
 
 
+def update_generation_info(args):
+    generation_info, html_info, img_index = args
+    try:
+        generation_info = json.loads(generation_info)
+        if img_index < 0 or img_index >= len(generation_info["infotexts"]):
+            return html_info
+        return plaintext_to_html(generation_info["infotexts"][img_index])
+    except Exception:
+        pass
+    # if the json parse or anything else fails, just return the old html_info
+    return html_info
+
+
 def create_refresh_button(refresh_component, refresh_method, refreshed_args, elem_id):
     def refresh():
         refresh_method()
@@ -656,6 +589,15 @@ Requested path was: {f}
                     with gr.Group():
                         html_info = gr.HTML()
                         generation_info = gr.Textbox(visible=False)
+                        if tabname == 'txt2img' or tabname == 'img2img':
+                            generation_info_button = gr.Button(visible=False, elem_id=f"{tabname}_generation_info_button")
+                            generation_info_button.click(
+                                fn=update_generation_info,
+                                _js="(x, y) => [x, y, selected_gallery_index()]",
+                                inputs=[generation_info, html_info],
+                                outputs=[html_info],
+                                preprocess=False
+                            )
 
                         save.click(
                             fn=wrap_gradio_call(save_files),
@@ -680,7 +622,7 @@ Requested path was: {f}
                 return result_gallery, generation_info if tabname != "extras" else html_info_x, html_info
 
 
-def create_ui(wrap_gradio_gpu_call):
+def create_ui():
     import modules.img2img
     import modules.txt2img
 
@@ -688,6 +630,9 @@ def create_ui(wrap_gradio_gpu_call):
 
     parameters_copypaste.reset()
 
+    modules.scripts.scripts_current = modules.scripts.scripts_txt2img
+    modules.scripts.scripts_txt2img.initialize_scripts(is_img2img=False)
+
     with gr.Blocks(analytics_enabled=False) as txt2img_interface:
         txt2img_prompt, roll, txt2img_prompt_style, txt2img_negative_prompt, txt2img_prompt_style2, submit, _, _,txt2img_prompt_style_apply, txt2img_save_style, txt2img_paste, token_counter, token_button, clear_prompt_button = create_toprow(is_img2img=False)
 
@@ -712,8 +657,8 @@ def create_ui(wrap_gradio_gpu_call):
                 sampler_index = gr.Radio(label='Sampling method', elem_id="txt2img_sampling", choices=[x.name for x in samplers], value=samplers[0].name, type="index")
 
                 with gr.Group():
-                    width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
-                    height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
+                    width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512)
+                    height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512)
 
                 with gr.Row():
                     restore_faces = gr.Checkbox(label='Restore faces', value=False, visible=len(shared.face_restorers) > 1)
@@ -721,8 +666,8 @@ def create_ui(wrap_gradio_gpu_call):
                     enable_hr = gr.Checkbox(label='Highres. fix', value=False)
 
                 with gr.Row(visible=False) as hr_options:
-                    firstphase_width = gr.Slider(minimum=0, maximum=1024, step=64, label="Firstpass width", value=0)
-                    firstphase_height = gr.Slider(minimum=0, maximum=1024, step=64, label="Firstpass height", value=0)
+                    firstphase_width = gr.Slider(minimum=0, maximum=1024, step=8, label="Firstpass width", value=0)
+                    firstphase_height = gr.Slider(minimum=0, maximum=1024, step=8, label="Firstpass height", value=0)
                     denoising_strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, label='Denoising strength', value=0.7)
 
                 with gr.Row(equal_height=True):
@@ -734,7 +679,7 @@ def create_ui(wrap_gradio_gpu_call):
                 seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs()
 
                 with gr.Group():
-                    custom_inputs = modules.scripts.scripts_txt2img.setup_ui(is_img2img=False)
+                    custom_inputs = modules.scripts.scripts_txt2img.setup_ui()
 
             txt2img_gallery, generation_info, html_info = create_output_panel("txt2img", opts.outdir_txt2img_samples)
             parameters_copypaste.bind_buttons({"txt2img": txt2img_paste}, None, txt2img_prompt)
@@ -842,7 +787,10 @@ def create_ui(wrap_gradio_gpu_call):
                 height,
             ]
 
-            token_button.click(fn=update_token_counter, inputs=[txt2img_prompt, steps], outputs=[token_counter])
+            token_button.click(fn=wrap_queued_call(update_token_counter), inputs=[txt2img_prompt, steps], outputs=[token_counter])
+
+    modules.scripts.scripts_current = modules.scripts.scripts_img2img
+    modules.scripts.scripts_img2img.initialize_scripts(is_img2img=True)
 
     with gr.Blocks(analytics_enabled=False) as img2img_interface:
         img2img_prompt, roll, img2img_prompt_style, img2img_negative_prompt, img2img_prompt_style2, submit, img2img_interrogate, img2img_deepbooru, img2img_prompt_style_apply, img2img_save_style, img2img_paste,token_counter, token_button, clear_prompt_button = create_toprow(is_img2img=True)
@@ -867,12 +815,26 @@ def create_ui(wrap_gradio_gpu_call):
                         init_img = gr.Image(label="Image for img2img", elem_id="img2img_image", show_label=False, source="upload", interactive=True, type="pil", tool=cmd_opts.gradio_img2img_tool).style(height=480)
 
                     with gr.TabItem('Inpaint', id='inpaint'):
-                        init_img_with_mask = gr.Image(label="Image for inpainting with mask",  show_label=False, elem_id="img2maskimg", source="upload", interactive=True, type="pil", tool="sketch", image_mode="RGBA").style(height=480)
+                        init_img_with_mask = gr.Image(label="Image for inpainting with mask", show_label=False, elem_id="img2maskimg", source="upload", interactive=True, type="pil", tool=cmd_opts.gradio_inpaint_tool, image_mode="RGBA").style(height=480)
+                        init_img_with_mask_orig = gr.State(None)
+
+                        use_color_sketch = cmd_opts.gradio_inpaint_tool == "color-sketch"
+                        if use_color_sketch:
+                            def update_orig(image, state):
+                                if image is not None:
+                                    same_size = state is not None and state.size == image.size
+                                    has_exact_match = np.any(np.all(np.array(image) == np.array(state), axis=-1))
+                                    edited = same_size and has_exact_match
+                                    return image if not edited or state is None else state
+
+                            init_img_with_mask.change(update_orig, [init_img_with_mask, init_img_with_mask_orig], init_img_with_mask_orig)
 
                         init_img_inpaint = gr.Image(label="Image for img2img", show_label=False, source="upload", interactive=True, type="pil", visible=False, elem_id="img_inpaint_base")
                         init_mask_inpaint = gr.Image(label="Mask", source="upload", interactive=True, type="pil", visible=False, elem_id="img_inpaint_mask")
 
-                        mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4)
+                        with gr.Row():
+                            mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4)
+                            mask_alpha = gr.Slider(label="Mask transparency", interactive=use_color_sketch, visible=use_color_sketch)
 
                         with gr.Row():
                             mask_mode = gr.Radio(label="Mask mode", show_label=False, choices=["Draw mask", "Upload mask"], type="index", value="Draw mask", elem_id="mask_mode")
@@ -897,8 +859,8 @@ def create_ui(wrap_gradio_gpu_call):
                 sampler_index = gr.Radio(label='Sampling method', choices=[x.name for x in samplers_for_img2img], value=samplers_for_img2img[0].name, type="index")
 
                 with gr.Group():
-                    width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512, elem_id="img2img_width")
-                    height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512, elem_id="img2img_height")
+                    width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="img2img_width")
+                    height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="img2img_height")
 
                 with gr.Row():
                     restore_faces = gr.Checkbox(label='Restore faces', value=False, visible=len(shared.face_restorers) > 1)
@@ -915,7 +877,7 @@ def create_ui(wrap_gradio_gpu_call):
                 seed, reuse_seed, subseed, reuse_subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w, seed_checkbox = create_seed_inputs()
 
                 with gr.Group():
-                    custom_inputs = modules.scripts.scripts_img2img.setup_ui(is_img2img=True)
+                    custom_inputs = modules.scripts.scripts_img2img.setup_ui()
 
             img2img_gallery, generation_info, html_info = create_output_panel("img2img", opts.outdir_img2img_samples)
             parameters_copypaste.bind_buttons({"img2img": img2img_paste}, None, img2img_prompt)
@@ -960,12 +922,14 @@ def create_ui(wrap_gradio_gpu_call):
                     img2img_prompt_style2,
                     init_img,
                     init_img_with_mask,
+                    init_img_with_mask_orig,
                     init_img_inpaint,
                     init_mask_inpaint,
                     mask_mode,
                     steps,
                     sampler_index,
                     mask_blur,
+                    mask_alpha,
                     inpainting_fill,
                     restore_faces,
                     tiling,
@@ -1001,11 +965,10 @@ def create_ui(wrap_gradio_gpu_call):
                 outputs=[img2img_prompt],
             )
 
-            if cmd_opts.deepdanbooru:
-                img2img_deepbooru.click(
-                    fn=interrogate_deepbooru,
-                    inputs=[init_img],
-                    outputs=[img2img_prompt],
+            img2img_deepbooru.click(
+                fn=interrogate_deepbooru,
+                inputs=[init_img],
+                outputs=[img2img_prompt],
             )
 
 
@@ -1060,11 +1023,14 @@ def create_ui(wrap_gradio_gpu_call):
                 (seed_resize_from_w, "Seed resize from-1"),
                 (seed_resize_from_h, "Seed resize from-2"),
                 (denoising_strength, "Denoising strength"),
+                (mask_blur, "Mask blur"),
                 *modules.scripts.scripts_img2img.infotext_fields
             ]
             parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields)
             parameters_copypaste.add_paste_fields("inpaint", init_img_with_mask, img2img_paste_fields)
 
+    modules.scripts.scripts_current = None
+
     with gr.Blocks(analytics_enabled=False) as extras_interface:
         with gr.Row().style(equal_height=False):
             with gr.Column(variant='panel'):
@@ -1178,7 +1144,11 @@ def create_ui(wrap_gradio_gpu_call):
                 custom_name = gr.Textbox(label="Custom Name (Optional)")
                 interp_amount = gr.Slider(minimum=0.0, maximum=1.0, step=0.05, label='Multiplier (M) - set to 0 to get model A', value=0.3)
                 interp_method = gr.Radio(choices=["Weighted sum", "Add difference"], value="Weighted sum", label="Interpolation Method")
-                save_as_half = gr.Checkbox(value=False, label="Save as float16")
+
+                with gr.Row():
+                    checkpoint_format = gr.Radio(choices=["ckpt", "safetensors"], value="ckpt", label="Checkpoint format")
+                    save_as_half = gr.Checkbox(value=False, label="Save as float16")
+
                 modelmerger_merge = gr.Button(elem_id="modelmerger_merge", label="Merge", variant='primary')
 
             with gr.Column(variant='panel'):
@@ -1208,7 +1178,7 @@ def create_ui(wrap_gradio_gpu_call):
 
                 with gr.Tab(label="Create hypernetwork"):
                     new_hypernetwork_name = gr.Textbox(label="Name")
-                    new_hypernetwork_sizes = gr.CheckboxGroup(label="Modules", value=["768", "320", "640", "1280"], choices=["768", "320", "640", "1280"])
+                    new_hypernetwork_sizes = gr.CheckboxGroup(label="Modules", value=["768", "320", "640", "1280"], choices=["768", "1024", "320", "640", "1280"])
                     new_hypernetwork_layer_structure = gr.Textbox("1, 2, 1", label="Enter hypernetwork layer structure", placeholder="1st and last digit must be 1. ex:'1, 2, 1'")
                     new_hypernetwork_activation_func = gr.Dropdown(value="linear", label="Select activation function of hypernetwork. Recommended : Swish / Linear(none)", choices=modules.hypernetworks.ui.keys)
                     new_hypernetwork_initialization_option = gr.Dropdown(value = "Normal", label="Select Layer weights initialization. Recommended: Kaiming for relu-like, Xavier for sigmoid-like, Normal otherwise", choices=["Normal", "KaimingUniform", "KaimingNormal", "XavierUniform", "XavierNormal"])
@@ -1226,8 +1196,8 @@ def create_ui(wrap_gradio_gpu_call):
                 with gr.Tab(label="Preprocess images"):
                     process_src = gr.Textbox(label='Source directory')
                     process_dst = gr.Textbox(label='Destination directory')
-                    process_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
-                    process_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
+                    process_width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512)
+                    process_height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512)
                     preprocess_txt_action = gr.Dropdown(label='Existing Caption txt Action', value="ignore", choices=["ignore", "copy", "prepend", "append"])
 
                     with gr.Row():
@@ -1235,7 +1205,7 @@ def create_ui(wrap_gradio_gpu_call):
                         process_split = gr.Checkbox(label='Split oversized images')
                         process_focal_crop = gr.Checkbox(label='Auto focal point crop')
                         process_caption = gr.Checkbox(label='Use BLIP for caption')
-                        process_caption_deepbooru = gr.Checkbox(label='Use deepbooru for caption', visible=True if cmd_opts.deepdanbooru else False)
+                        process_caption_deepbooru = gr.Checkbox(label='Use deepbooru for caption', visible=True)
 
                     with gr.Row(visible=False) as process_split_extra_row:
                         process_split_threshold = gr.Slider(label='Split image threshold', value=0.5, minimum=0.0, maximum=1.0, step=0.05)
@@ -1252,6 +1222,8 @@ def create_ui(wrap_gradio_gpu_call):
                             gr.HTML(value="")
 
                         with gr.Column():
+                            with gr.Row():
+                                interrupt_preprocessing = gr.Button("Interrupt")
                             run_preprocess = gr.Button(value="Preprocess", variant='primary')
 
                     process_split.change(
@@ -1279,22 +1251,32 @@ def create_ui(wrap_gradio_gpu_call):
                         hypernetwork_learn_rate = gr.Textbox(label='Hypernetwork Learning rate', placeholder="Hypernetwork Learning rate", value="0.00001")
 
                     batch_size = gr.Number(label='Batch size', value=1, precision=0)
+                    gradient_step = gr.Number(label='Gradient accumulation steps', value=1, precision=0)
                     dataset_directory = gr.Textbox(label='Dataset directory', placeholder="Path to directory with input images")
                     log_directory = gr.Textbox(label='Log directory', placeholder="Path to directory where to write outputs", value="textual_inversion")
                     template_file = gr.Textbox(label='Prompt template file', value=os.path.join(script_path, "textual_inversion_templates", "style_filewords.txt"))
-                    training_width = gr.Slider(minimum=64, maximum=2048, step=64, label="Width", value=512)
-                    training_height = gr.Slider(minimum=64, maximum=2048, step=64, label="Height", value=512)
+                    training_width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512)
+                    training_height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512)
                     steps = gr.Number(label='Max steps', value=100000, precision=0)
                     create_image_every = gr.Number(label='Save an image to log directory every N steps, 0 to disable', value=500, precision=0)
                     save_embedding_every = gr.Number(label='Save a copy of embedding to log directory every N steps, 0 to disable', value=500, precision=0)
                     save_image_with_stored_embedding = gr.Checkbox(label='Save images with embedding in PNG chunks', value=True)
                     preview_from_txt2img = gr.Checkbox(label='Read parameters (prompt, etc...) from txt2img tab when making previews', value=False)
+                    with gr.Row():
+                        shuffle_tags = gr.Checkbox(label="Shuffle tags by ',' when creating prompts.", value=False)
+                        tag_drop_out = gr.Slider(minimum=0, maximum=1, step=0.1, label="Drop out tags when creating prompts.", value=0)
+                    with gr.Row():
+                        latent_sampling_method = gr.Radio(label='Choose latent sampling method', value="once", choices=['once', 'deterministic', 'random'])
 
                     with gr.Row():
                         interrupt_training = gr.Button(value="Interrupt")
                         train_hypernetwork = gr.Button(value="Train Hypernetwork", variant='primary')
                         train_embedding = gr.Button(value="Train Embedding", variant='primary')
 
+                params = script_callbacks.UiTrainTabParams(txt2img_preview_params)
+
+                script_callbacks.ui_train_tabs_callback(params)
+
             with gr.Column():
                 progressbar = gr.HTML(elem_id="ti_progressbar")
                 ti_output = gr.Text(elem_id="ti_output", value="", show_label=False)
@@ -1373,11 +1355,15 @@ def create_ui(wrap_gradio_gpu_call):
                 train_embedding_name,
                 embedding_learn_rate,
                 batch_size,
+                gradient_step,
                 dataset_directory,
                 log_directory,
                 training_width,
                 training_height,
                 steps,
+                shuffle_tags,
+                tag_drop_out,
+                latent_sampling_method,
                 create_image_every,
                 save_embedding_every,
                 template_file,
@@ -1398,11 +1384,15 @@ def create_ui(wrap_gradio_gpu_call):
                 train_hypernetwork_name,
                 hypernetwork_learn_rate,
                 batch_size,
+                gradient_step,
                 dataset_directory,
                 log_directory,
                 training_width,
                 training_height,
                 steps,
+                shuffle_tags,
+                tag_drop_out,
+                latent_sampling_method,
                 create_image_every,
                 save_embedding_every,
                 template_file,
@@ -1421,6 +1411,12 @@ def create_ui(wrap_gradio_gpu_call):
             outputs=[],
         )
 
+        interrupt_preprocessing.click(
+            fn=lambda: shared.state.interrupt(),
+            inputs=[],
+            outputs=[],
+        )
+
     def create_setting_component(key, is_quicksettings=False):
         def fun():
             return opts.data[key] if key in opts.data else opts.data_labels[key].default
@@ -1472,35 +1468,21 @@ def create_ui(wrap_gradio_gpu_call):
             if comp == dummy_component:
                 continue
 
-            oldval = opts.data.get(key, None)
-            try:
-                setattr(opts, key, value)
-            except RuntimeError:
-                continue
-            if oldval != value:
-                if opts.data_labels[key].onchange is not None:
-                    opts.data_labels[key].onchange()
-
+            if opts.set(key, value):
                 changed.append(key)
+
         try:
             opts.save(shared.config_filename)
         except RuntimeError:
             return opts.dumpjson(), f'{len(changed)} settings changed without save: {", ".join(changed)}.'
-        return opts.dumpjson(), f'{len(changed)} settings changed: {", ".join(changed)}.'
+        return opts.dumpjson(), f'{len(changed)} settings changed{": " if len(changed) > 0 else ""}{", ".join(changed)}.'
 
     def run_settings_single(value, key):
         if not opts.same_type(value, opts.data_labels[key].default):
             return gr.update(visible=True), opts.dumpjson()
 
-        oldval = opts.data.get(key, None)
-        try:
-            setattr(opts, key, value)
-        except Exception:
-            return gr.update(value=oldval), opts.dumpjson()
-
-        if oldval != value:
-            if opts.data_labels[key].onchange is not None:
-                opts.data_labels[key].onchange()
+        if not opts.set(key, value):
+            return gr.update(value=getattr(opts, key)), opts.dumpjson()
 
         opts.save(shared.config_filename)
 
@@ -1694,6 +1676,7 @@ def create_ui(wrap_gradio_gpu_call):
                 interp_amount,
                 save_as_half,
                 custom_name,
+                checkpoint_format,
             ],
             outputs=[
                 submit_result,
@@ -1777,7 +1760,7 @@ def create_ui(wrap_gradio_gpu_call):
     return demo
 
 
-def load_javascript(raw_response):
+def reload_javascript():
     with open(os.path.join(script_path, "script.js"), "r", encoding="utf8") as jsfile:
         javascript = f'<script>{jsfile.read()}</script>'
 
@@ -1793,7 +1776,7 @@ def load_javascript(raw_response):
     javascript += f"\n<script>{localization.localization_js(shared.opts.localization)}</script>"
 
     def template_response(*args, **kwargs):
-        res = raw_response(*args, **kwargs)
+        res = shared.GradioTemplateResponseOriginal(*args, **kwargs)
         res.body = res.body.replace(
             b'</head>', f'{javascript}</head>'.encode("utf8"))
         res.init_headers()
@@ -1802,4 +1785,5 @@ def load_javascript(raw_response):
     gradio.routes.templates.TemplateResponse = template_response
 
 
-reload_javascript = partial(load_javascript, gradio.routes.templates.TemplateResponse)
+if not hasattr(shared, 'GradioTemplateResponseOriginal'):
+    shared.GradioTemplateResponseOriginal = gradio.routes.templates.TemplateResponse

modules/ui_extensions.py

@@ -17,7 +17,7 @@ available_extensions = {"extensions": []}
 
 
 def check_access():
-    assert not shared.cmd_opts.disable_extension_access, "extension access disabed because of commandline flags"
+    assert not shared.cmd_opts.disable_extension_access, "extension access disabled because of command line flags"
 
 
 def apply_and_restart(disable_list, update_list):
@@ -36,9 +36,9 @@ def apply_and_restart(disable_list, update_list):
             continue
 
         try:
-            ext.pull()
+            ext.fetch_and_reset_hard()
         except Exception:
-            print(f"Error pulling updates for {ext.name}:", file=sys.stderr)
+            print(f"Error getting updates for {ext.name}:", file=sys.stderr)
             print(traceback.format_exc(), file=sys.stderr)
 
     shared.opts.disabled_extensions = disabled
@@ -78,6 +78,12 @@ def extension_table():
     """
 
     for ext in extensions.extensions:
+        remote = ""
+        if ext.is_builtin:
+            remote = "built-in"
+        elif ext.remote:
+            remote = f"""<a href="{html.escape(ext.remote or '')}" target="_blank">{html.escape("built-in" if ext.is_builtin else ext.remote or '')}</a>"""
+
         if ext.can_update:
             ext_status = f"""<label><input class="gr-check-radio gr-checkbox" name="update_{html.escape(ext.name)}" checked="checked" type="checkbox">{html.escape(ext.status)}</label>"""
         else:
@@ -86,7 +92,7 @@ def extension_table():
         code += f"""
             <tr>
                 <td><label><input class="gr-check-radio gr-checkbox" name="enable_{html.escape(ext.name)}" type="checkbox" {'checked="checked"' if ext.enabled else ''}>{html.escape(ext.name)}</label></td>
-                <td><a href="{html.escape(ext.remote or '')}" target="_blank">{html.escape(ext.remote or '')}</a></td>
+                <td>{remote}</td>
                 <td{' class="extension_status"' if ext.remote is not None else ''}>{ext_status}</td>
             </tr>
     """
@@ -134,6 +140,9 @@ def install_extension_from_url(dirname, url):
 
         os.rename(tmpdir, target_dir)
 
+        import launch
+        launch.run_extension_installer(target_dir)
+
         extensions.list_extensions()
         return [extension_table(), html.escape(f"Installed into {target_dir}. Use Installed tab to restart.")]
     finally:

modules/ui_tempdir.py

@@ -0,0 +1,62 @@
+import os
+import tempfile
+from collections import namedtuple
+
+import gradio as gr
+
+from PIL import PngImagePlugin
+
+from modules import shared
+
+
+Savedfile = namedtuple("Savedfile", ["name"])
+
+
+def save_pil_to_file(pil_image, dir=None):
+    already_saved_as = getattr(pil_image, 'already_saved_as', None)
+    if already_saved_as and os.path.isfile(already_saved_as):
+        shared.demo.temp_dirs = shared.demo.temp_dirs | {os.path.abspath(os.path.dirname(already_saved_as))}
+        file_obj = Savedfile(already_saved_as)
+        return file_obj
+
+    if shared.opts.temp_dir != "":
+        dir = shared.opts.temp_dir
+
+    use_metadata = False
+    metadata = PngImagePlugin.PngInfo()
+    for key, value in pil_image.info.items():
+        if isinstance(key, str) and isinstance(value, str):
+            metadata.add_text(key, value)
+            use_metadata = True
+
+    file_obj = tempfile.NamedTemporaryFile(delete=False, suffix=".png", dir=dir)
+    pil_image.save(file_obj, pnginfo=(metadata if use_metadata else None))
+    return file_obj
+
+
+# override save to file function so that it also writes PNG info
+gr.processing_utils.save_pil_to_file = save_pil_to_file
+
+
+def on_tmpdir_changed():
+    if shared.opts.temp_dir == "" or shared.demo is None:
+        return
+
+    os.makedirs(shared.opts.temp_dir, exist_ok=True)
+
+    shared.demo.temp_dirs = shared.demo.temp_dirs | {os.path.abspath(shared.opts.temp_dir)}
+
+
+def cleanup_tmpdr():
+    temp_dir = shared.opts.temp_dir
+    if temp_dir == "" or not os.path.isdir(temp_dir):
+        return
+
+    for root, dirs, files in os.walk(temp_dir, topdown=False):
+        for name in files:
+            _, extension = os.path.splitext(name)
+            if extension != ".png":
+                continue
+
+            filename = os.path.join(root, name)
+            os.remove(filename)

requirements.txt

@@ -1,3 +1,4 @@
+accelerate
 basicsr
 diffusers
 fairscale==0.4.4
@@ -27,3 +28,6 @@ kornia
 lark
 inflection
 GitPython
+torchsde
+safetensors
+psutil; sys_platform == 'darwin'

requirements_versions.txt

@@ -1,5 +1,6 @@
 transformers==4.19.2
 diffusers==0.3.0
+accelerate==0.12.0
 basicsr==1.4.2
 gfpgan==1.3.8
 gradio==3.9
@@ -24,3 +25,5 @@ kornia==0.6.7
 lark==1.1.2
 inflection==0.5.1
 GitPython==3.1.27
+torchsde==0.2.5
+safetensors==0.2.5

scripts/img2imgalt.py

@@ -157,7 +157,7 @@ class Script(scripts.Script):
     def run(self, p, _, override_sampler, override_prompt, original_prompt, original_negative_prompt, override_steps, st, override_strength, cfg, randomness, sigma_adjustment):
         # Override
         if override_sampler:
-            p.sampler_index = [sampler.name for sampler in sd_samplers.samplers].index("Euler")
+            p.sampler_name = "Euler"
         if override_prompt:
             p.prompt = original_prompt
             p.negative_prompt = original_negative_prompt
@@ -191,7 +191,7 @@ class Script(scripts.Script):
             
             combined_noise = ((1 - randomness) * rec_noise + randomness * rand_noise) / ((randomness**2 + (1-randomness)**2) ** 0.5)
             
-            sampler = sd_samplers.create_sampler_with_index(sd_samplers.samplers, p.sampler_index, p.sd_model)
+            sampler = sd_samplers.create_sampler(p.sampler_name, p.sd_model)
 
             sigmas = sampler.model_wrap.get_sigmas(p.steps)
             

scripts/prompt_matrix.py

@@ -80,8 +80,10 @@ class Script(scripts.Script):
         grid = images.image_grid(processed.images, p.batch_size, rows=1 << ((len(prompt_matrix_parts) - 1) // 2))
         grid = images.draw_prompt_matrix(grid, p.width, p.height, prompt_matrix_parts)
         processed.images.insert(0, grid)
+        processed.index_of_first_image = 1
+        processed.infotexts.insert(0, processed.infotexts[0])
 
         if opts.grid_save:
-            images.save_image(processed.images[0], p.outpath_grids, "prompt_matrix", prompt=original_prompt, seed=processed.seed, grid=True, p=p)
+            images.save_image(processed.images[0], p.outpath_grids, "prompt_matrix", extension=opts.grid_format, prompt=original_prompt, seed=processed.seed, grid=True, p=p)
 
         return processed

scripts/prompts_from_file.py

@@ -145,6 +145,8 @@ class Script(scripts.Script):
         state.job_count = job_count
 
         images = []
+        all_prompts = []
+        infotexts = []
         for n, args in enumerate(jobs):
             state.job = f"{state.job_no + 1} out of {state.job_count}"
 
@@ -157,5 +159,7 @@ class Script(scripts.Script):
             
             if checkbox_iterate:
                 p.seed = p.seed + (p.batch_size * p.n_iter)
+            all_prompts += proc.all_prompts
+            infotexts += proc.infotexts
 
-        return Processed(p, images, p.seed, "")
+        return Processed(p, images, p.seed, "", all_prompts=all_prompts, infotexts=infotexts)

scripts/xy_grid.py

@@ -10,9 +10,9 @@ import numpy as np
 import modules.scripts as scripts
 import gradio as gr
 
-from modules import images
+from modules import images, sd_samplers
 from modules.hypernetworks import hypernetwork
-from modules.processing import process_images, Processed, get_correct_sampler, StableDiffusionProcessingTxt2Img
+from modules.processing import process_images, Processed, StableDiffusionProcessingTxt2Img
 from modules.shared import opts, cmd_opts, state
 import modules.shared as shared
 import modules.sd_samplers
@@ -58,29 +58,19 @@ def apply_order(p, x, xs):
         prompt_tmp += part
         prompt_tmp += x[idx]
     p.prompt = prompt_tmp + p.prompt
-    
-
-def build_samplers_dict(p):
-    samplers_dict = {}
-    for i, sampler in enumerate(get_correct_sampler(p)):
-        samplers_dict[sampler.name.lower()] = i
-        for alias in sampler.aliases:
-            samplers_dict[alias.lower()] = i
-    return samplers_dict
 
 
 def apply_sampler(p, x, xs):
-    sampler_index = build_samplers_dict(p).get(x.lower(), None)
-    if sampler_index is None:
+    sampler_name = sd_samplers.samplers_map.get(x.lower(), None)
+    if sampler_name is None:
         raise RuntimeError(f"Unknown sampler: {x}")
 
-    p.sampler_index = sampler_index
+    p.sampler_name = sampler_name
 
 
 def confirm_samplers(p, xs):
-    samplers_dict = build_samplers_dict(p)
     for x in xs:
-        if x.lower() not in samplers_dict.keys():
+        if x.lower() not in sd_samplers.samplers_map:
             raise RuntimeError(f"Unknown sampler: {x}")
 
 
@@ -393,6 +383,6 @@ class Script(scripts.Script):
             )
 
         if opts.grid_save:
-            images.save_image(processed.images[0], p.outpath_grids, "xy_grid", prompt=p.prompt, seed=processed.seed, grid=True, p=p)
+            images.save_image(processed.images[0], p.outpath_grids, "xy_grid", extension=opts.grid_format, prompt=p.prompt, seed=processed.seed, grid=True, p=p)
 
         return processed

test/advanced_features/extras_test.py

@@ -11,8 +11,8 @@ class TestExtrasWorking(unittest.TestCase):
             "codeformer_visibility": 0,
             "codeformer_weight": 0,
             "upscaling_resize": 2,
-            "upscaling_resize_w": 512,
-            "upscaling_resize_h": 512,
+            "upscaling_resize_w": 128,
+            "upscaling_resize_h": 128,
             "upscaling_crop": True,
             "upscaler_1": "None",
             "upscaler_2": "None",

test/advanced_features/txt2img_test.py

@@ -0,0 +1,47 @@
+import unittest
+import requests
+
+
+class TestTxt2ImgWorking(unittest.TestCase):
+    def setUp(self):
+        self.url_txt2img = "http://localhost:7860/sdapi/v1/txt2img"
+        self.simple_txt2img = {
+            "enable_hr": False,
+            "denoising_strength": 0,
+            "firstphase_width": 0,
+            "firstphase_height": 0,
+            "prompt": "example prompt",
+            "styles": [],
+            "seed": -1,
+            "subseed": -1,
+            "subseed_strength": 0,
+            "seed_resize_from_h": -1,
+            "seed_resize_from_w": -1,
+            "batch_size": 1,
+            "n_iter": 1,
+            "steps": 3,
+            "cfg_scale": 7,
+            "width": 64,
+            "height": 64,
+            "restore_faces": False,
+            "tiling": False,
+            "negative_prompt": "",
+            "eta": 0,
+            "s_churn": 0,
+            "s_tmax": 0,
+            "s_tmin": 0,
+            "s_noise": 1,
+            "sampler_index": "Euler a"
+        }
+
+    def test_txt2img_with_restore_faces_performed(self):
+        self.simple_txt2img["restore_faces"] = True
+        self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
+
+
+class TestTxt2ImgCorrectness(unittest.TestCase):
+    pass
+
+
+if __name__ == "__main__":
+    unittest.main()

test/basic_features/img2img_test.py

@@ -51,9 +51,5 @@ class TestImg2ImgWorking(unittest.TestCase):
         self.assertEqual(requests.post(self.url_img2img, json=self.simple_img2img).status_code, 200)
 
 
-class TestImg2ImgCorrectness(unittest.TestCase):
-    pass
-
-
 if __name__ == "__main__":
     unittest.main()

test/basic_features/txt2img_test.py

@@ -49,26 +49,20 @@ class TestTxt2ImgWorking(unittest.TestCase):
         self.simple_txt2img["enable_hr"] = True
         self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
 
-    def test_txt2img_with_restore_faces_performed(self):
-        self.simple_txt2img["restore_faces"] = True
-        self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
-
-    def test_txt2img_with_tiling_faces_performed(self):
+    def test_txt2img_with_tiling_performed(self):
         self.simple_txt2img["tiling"] = True
         self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
 
     def test_txt2img_with_vanilla_sampler_performed(self):
         self.simple_txt2img["sampler_index"] = "PLMS"
         self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
+        self.simple_txt2img["sampler_index"] = "DDIM"
+        self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
 
     def test_txt2img_multiple_batches_performed(self):
         self.simple_txt2img["n_iter"] = 2
         self.assertEqual(requests.post(self.url_txt2img, json=self.simple_txt2img).status_code, 200)
 
 
-class TestTxt2ImgCorrectness(unittest.TestCase):
-    pass
-
-
 if __name__ == "__main__":
     unittest.main()

test/basic_features/utils_test.py

@@ -18,20 +18,6 @@ class UtilsTests(unittest.TestCase):
   def test_options_get(self):
     self.assertEqual(requests.get(self.url_options).status_code, 200)
 
-  def test_options_write(self):
-    response = requests.get(self.url_options)
-    self.assertEqual(response.status_code, 200)
-    
-    pre_value = response.json()["send_seed"]
-
-    self.assertEqual(requests.post(self.url_options, json={"send_seed":not pre_value}).status_code, 200)
-
-    response = requests.get(self.url_options)
-    self.assertEqual(response.status_code, 200)
-    self.assertEqual(response.json()["send_seed"], not pre_value)
-
-    requests.post(self.url_options, json={"send_seed": pre_value})
-
   def test_cmd_flags(self):
     self.assertEqual(requests.get(self.url_cmd_flags).status_code, 200)
 
@@ -60,4 +46,8 @@ class UtilsTests(unittest.TestCase):
     self.assertEqual(requests.get(self.url_artist_categories).status_code, 200)
 
   def test_artists(self):
-    self.assertEqual(requests.get(self.url_artists).status_code, 200)
+    self.assertEqual(requests.get(self.url_artists).status_code, 200)
+
+
+if __name__ == "__main__":
+    unittest.main()

test/server_poll.py

@@ -3,7 +3,7 @@ import requests
 import time
 
 
-def run_tests():
+def run_tests(proc, test_dir):
     timeout_threshold = 240
     start_time = time.time()
     while time.time()-start_time < timeout_threshold:
@@ -11,9 +11,14 @@ def run_tests():
             requests.head("http://localhost:7860/")
             break
         except requests.exceptions.ConnectionError:
-            pass
-    if time.time()-start_time < timeout_threshold:
-        suite = unittest.TestLoader().discover('', pattern='*_test.py')
+            if proc.poll() is not None:
+                break
+    if proc.poll() is None:
+        if test_dir is None:
+            test_dir = ""
+        suite = unittest.TestLoader().discover(test_dir, pattern="*_test.py", top_level_dir="test")
         result = unittest.TextTestRunner(verbosity=2).run(suite)
+        return len(result.failures) + len(result.errors)
     else:
         print("Launch unsuccessful")
+        return 1

v1-inference.yaml

@@ -0,0 +1,70 @@
+model:
+  base_learning_rate: 1.0e-04
+  target: ldm.models.diffusion.ddpm.LatentDiffusion
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "txt"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false   # Note: different from the one we trained before
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+
+    scheduler_config: # 10000 warmup steps
+      target: ldm.lr_scheduler.LambdaLinearScheduler
+      params:
+        warm_up_steps: [ 10000 ]
+        cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
+        f_start: [ 1.e-6 ]
+        f_max: [ 1. ]
+        f_min: [ 1. ]
+
+    unet_config:
+      target: ldm.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_heads: 8
+        use_spatial_transformer: True
+        transformer_depth: 1
+        context_dim: 768
+        use_checkpoint: True
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenCLIPEmbedder

webui-macos-env.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+####################################################################
+#                          macOS defaults                          #
+# Please modify webui-user.sh to change these instead of this file #
+####################################################################
+
+if [[ -x "$(command -v python3.10)" ]]
+then
+    python_cmd="python3.10"
+fi
+
+export install_dir="$HOME"
+export COMMANDLINE_ARGS="--skip-torch-cuda-test --no-half --use-cpu interrogate"
+export TORCH_COMMAND="pip install torch==1.12.1 torchvision==0.13.1"
+export K_DIFFUSION_REPO="https://github.com/brkirch/k-diffusion.git"
+export K_DIFFUSION_COMMIT_HASH="51c9778f269cedb55a4d88c79c0246d35bdadb71"
+export PYTORCH_ENABLE_MPS_FALLBACK=1
+
+####################################################################

webui-user.sh

@@ -10,7 +10,7 @@
 #clone_dir="stable-diffusion-webui"
 
 # Commandline arguments for webui.py, for example: export COMMANDLINE_ARGS="--medvram --opt-split-attention"
-export COMMANDLINE_ARGS=""
+#export COMMANDLINE_ARGS=""
 
 # python3 executable
 #python_cmd="python3"
@@ -40,4 +40,7 @@ export COMMANDLINE_ARGS=""
 #export CODEFORMER_COMMIT_HASH=""
 #export BLIP_COMMIT_HASH=""
 
+# Uncomment to enable accelerated launch
+#export ACCELERATE="True"
+
 ###########################################

webui.bat

@@ -28,15 +28,27 @@ goto :show_stdout_stderr
 :activate_venv
 set PYTHON="%~dp0%VENV_DIR%\Scripts\Python.exe"
 echo venv %PYTHON%
+if [%ACCELERATE%] == ["True"] goto :accelerate
 goto :launch
 
 :skip_venv
 
+:accelerate
+echo "Checking for accelerate"
+set ACCELERATE="%~dp0%VENV_DIR%\Scripts\accelerate.exe"
+if EXIST %ACCELERATE% goto :accelerate_launch
+
 :launch
 %PYTHON% launch.py %*
 pause
 exit /b
 
+:accelerate_launch
+echo "Accelerating"
+%ACCELERATE% launch --num_cpu_threads_per_process=6 launch.py
+pause
+exit /b
+
 :show_stdout_stderr
 
 echo.

webui.py

@@ -8,9 +8,10 @@ from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.middleware.gzip import GZipMiddleware
 
+from modules.call_queue import wrap_queued_call, queue_lock, wrap_gradio_gpu_call
 from modules.paths import script_path
 
-from modules import devices, sd_samplers, upscaler, extensions, localization
+from modules import shared, devices, sd_samplers, upscaler, extensions, localization, ui_tempdir
 import modules.codeformer_model as codeformer
 import modules.extras
 import modules.face_restoration
@@ -23,7 +24,6 @@ import modules.scripts
 import modules.sd_hijack
 import modules.sd_models
 import modules.sd_vae
-import modules.shared as shared
 import modules.txt2img
 import modules.script_callbacks
 
@@ -32,32 +32,11 @@ from modules import modelloader
 from modules.shared import cmd_opts
 import modules.hypernetworks.hypernetwork
 
-queue_lock = threading.Lock()
-server_name = "0.0.0.0" if cmd_opts.listen else cmd_opts.server_name
 
-def wrap_queued_call(func):
-    def f(*args, **kwargs):
-        with queue_lock:
-            res = func(*args, **kwargs)
-
-        return res
-
-    return f
-
-
-def wrap_gradio_gpu_call(func, extra_outputs=None):
-    def f(*args, **kwargs):
-
-        shared.state.begin()
-
-        with queue_lock:
-            res = func(*args, **kwargs)
-
-        shared.state.end()
-
-        return res
-
-    return modules.ui.wrap_gradio_call(f, extra_outputs=extra_outputs, add_stats=True)
+if cmd_opts.server_name:
+    server_name = cmd_opts.server_name
+else:
+    server_name = "0.0.0.0" if cmd_opts.listen else None
 
 
 def initialize():
@@ -74,16 +53,19 @@ def initialize():
     codeformer.setup_model(cmd_opts.codeformer_models_path)
     gfpgan.setup_model(cmd_opts.gfpgan_models_path)
     shared.face_restorers.append(modules.face_restoration.FaceRestoration())
-    modelloader.load_upscalers()
 
     modules.scripts.load_scripts()
 
+    modelloader.load_upscalers()
+
     modules.sd_vae.refresh_vae_list()
     modules.sd_models.load_model()
     shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: modules.sd_models.reload_model_weights()))
     shared.opts.onchange("sd_vae", wrap_queued_call(lambda: modules.sd_vae.reload_vae_weights()), call=False)
-    shared.opts.onchange("sd_hypernetwork", wrap_queued_call(lambda: modules.hypernetworks.hypernetwork.load_hypernetwork(shared.opts.sd_hypernetwork)))
+    shared.opts.onchange("sd_vae_as_default", wrap_queued_call(lambda: modules.sd_vae.reload_vae_weights()), call=False)
+    shared.opts.onchange("sd_hypernetwork", wrap_queued_call(lambda: shared.reload_hypernetworks()))
     shared.opts.onchange("sd_hypernetwork_strength", modules.hypernetworks.hypernetwork.apply_strength)
+    shared.opts.onchange("temp_dir", ui_tempdir.on_tmpdir_changed)
 
     if cmd_opts.tls_keyfile is not None and cmd_opts.tls_keyfile is not None:
 
@@ -107,8 +89,12 @@ def initialize():
 
 
 def setup_cors(app):
-    if cmd_opts.cors_allow_origins:
+    if cmd_opts.cors_allow_origins and cmd_opts.cors_allow_origins_regex:
+        app.add_middleware(CORSMiddleware, allow_origins=cmd_opts.cors_allow_origins.split(','), allow_origin_regex=cmd_opts.cors_allow_origins_regex, allow_methods=['*'])
+    elif cmd_opts.cors_allow_origins:
         app.add_middleware(CORSMiddleware, allow_origins=cmd_opts.cors_allow_origins.split(','), allow_methods=['*'])
+    elif cmd_opts.cors_allow_origins_regex:
+        app.add_middleware(CORSMiddleware, allow_origin_regex=cmd_opts.cors_allow_origins_regex, allow_methods=['*'])
 
 
 def create_api(app):
@@ -146,9 +132,12 @@ def webui():
     initialize()
 
     while 1:
-        demo = modules.ui.create_ui(wrap_gradio_gpu_call=wrap_gradio_gpu_call)
+        if shared.opts.clean_temp_dir_at_start:
+            ui_tempdir.cleanup_tmpdr()
+
+        shared.demo = modules.ui.create_ui()
 
-        app, local_url, share_url = demo.launch(
+        app, local_url, share_url = shared.demo.queue(default_enabled=False).launch(
             share=cmd_opts.share,
             server_name=server_name,
             server_port=cmd_opts.port,
@@ -175,9 +164,10 @@ def webui():
         if launch_api:
             create_api(app)
 
-        modules.script_callbacks.app_started_callback(demo, app)
+        modules.script_callbacks.app_started_callback(shared.demo, app)
+        modules.script_callbacks.app_started_callback(shared.demo, app)
 
-        wait_on_server(demo)
+        wait_on_server(shared.demo)
 
         sd_samplers.set_samplers()
 
@@ -188,6 +178,8 @@ def webui():
 
         print('Reloading custom scripts')
         modules.scripts.reload_scripts()
+        modelloader.load_upscalers()
+
         print('Reloading modules: modules.ui')
         importlib.reload(modules.ui)
         print('Refreshing Model List')

webui.sh

@@ -3,6 +3,15 @@
 # Please do not make any changes to this file,  #
 # change the variables in webui-user.sh instead #
 #################################################
+
+# If run from macOS, load defaults from webui-macos-env.sh
+if [[ "$OSTYPE" == "darwin"* ]]; then
+    if [[ -f webui-macos-env.sh ]]
+        then
+        source ./webui-macos-env.sh
+    fi
+fi
+
 # Read variables from webui-user.sh
 # shellcheck source=/dev/null
 if [[ -f webui-user.sh ]]
@@ -46,6 +55,18 @@ then
     LAUNCH_SCRIPT="launch.py"
 fi
 
+# this script cannot be run as root by default
+can_run_as_root=0
+
+# read any command line flags to the webui.sh script
+while getopts "f" flag > /dev/null 2>&1
+do
+    case ${flag} in
+        f) can_run_as_root=1;;
+        *) break;;
+    esac
+done
+
 # Disable sentry logging
 export ERROR_REPORTING=FALSE
 
@@ -61,7 +82,7 @@ printf "\e[1m\e[34mTested on Debian 11 (Bullseye)\e[0m"
 printf "\n%s\n" "${delimiter}"
 
 # Do not run as root
-if [[ $(id -u) -eq 0 ]]
+if [[ $(id -u) -eq 0 && can_run_as_root -eq 0 ]]
 then
     printf "\n%s\n" "${delimiter}"
     printf "\e[1m\e[31mERROR: This script must not be launched as root, aborting...\e[0m"
@@ -134,7 +155,15 @@ else
     exit 1
 fi
 
-printf "\n%s\n" "${delimiter}"
-printf "Launching launch.py..."
-printf "\n%s\n" "${delimiter}"
-"${python_cmd}" "${LAUNCH_SCRIPT}" "$@"
+if [[ ! -z "${ACCELERATE}" ]] && [ ${ACCELERATE}="True" ] && [ -x "$(command -v accelerate)" ]
+then
+    printf "\n%s\n" "${delimiter}"
+    printf "Accelerating launch.py..."
+    printf "\n%s\n" "${delimiter}"
+    accelerate launch --num_cpu_threads_per_process=6 "${LAUNCH_SCRIPT}" "$@"
+else
+    printf "\n%s\n" "${delimiter}"
+    printf "Launching launch.py..."
+    printf "\n%s\n" "${delimiter}"
+    "${python_cmd}" "${LAUNCH_SCRIPT}" "$@"
+fi