added detailed installation instructions

fixed bug with missing same dir for a new install
added ctrl+c hander to immediately stop the program instead of waiting

README.md

@@ -6,50 +6,77 @@ Original script with Gradio UI was written by a kind anonymous user. This is a m
 ![](screenshot.png)
 ## Installing and running
 
-### Stable Diffusion
+You need python and git installed to run this. I tested the installation to work with Python 3.8.10,
+you may be able to run this on different versions.
 
-This script assumes that you already have main Stable Diffusion sutff installed, assumed to be in directory `/sd`.
+You need Stable Diffusion model checkpoint, a big file containing the neural network weights. You
-If you don't have it installed, follow the guide:
+can obtain it from the following places:
+ - [official download](https://huggingface.co/CompVis/stable-diffusion-v-1-4-original)
+ - [file storage](https://drive.yerf.org/wl/?id=EBfTrmcCCUAGaQBXVIj5lJmEhjoP1tgl)
+ - [torrent](magnet:?xt=urn:btih:3a4a612d75ed088ea542acac52f9f45987488d1c&dn=sd-v1-4.ckpt&tr=udp%3a%2f%2ftracker.openbittorrent.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.opentrackr.org%3a1337)
 
-- https://rentry.org/kretard
+You optionally can use GPFGAN to improve faces, then you'll need to download the model from [here](https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth).
 
-This repository's `webgui.py` is a replacement for `kdiff.py` from the guide.
+Instructions:
 
-Particularly, following files must exist:
+```commandline
+:: crate a directory somewhere for stable diffusion and open cmd in it; below the directorty is assumed to be b:\src\sd
+:: make sure you are in the right directory; the command must output b:\src\sd1
+echo %cd%
+
+:: install torch with CUDA support. See https://pytorch.org/get-started/locally/ for more instructions if this fails.
+pip install torch --extra-index-url https://download.pytorch.org/whl/cu113
+
+:: check if torch supports GPU; this must output "True". You need CUDA 11. installed for this. You might be able to use
+:: a different version, but this is what I tested.
+python -c "import torch; print(torch.cuda.is_available())"
 
-- `/sd/configs/stable-diffusion/v1-inference.yaml`
+:: clone Stable Diffusion repositories
-- `/sd/models/ldm/stable-diffusion-v1/model.ckpt`
+git clone https://github.com/CompVis/stable-diffusion.git
-- `/sd/ldm/util.py`
+git clone https://github.com/CompVis/taming-transformers
-- `/sd/k_diffusion/__init__.py`
 
-### GFPGAN
+:: install requirements of Stable Diffusion
+pip install transformers==4.19.2 diffusers invisible-watermark
 
-If you want to use GFPGAN to improve generated faces, you need to install it separately.
+:: install k-diffusion
-Follow instructions from https://github.com/TencentARC/GFPGAN, but when cloning it, do so into Stable Diffusion main directory, `/sd`.
+pip install git+https://github.com/crowsonkb/k-diffusion.git
-After that download [GFPGANv1.3.pth](https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth) and put it
-into the `/sd/GFPGAN/experiments/pretrained_models` directory. If you're getting troubles with GFPGAN support, follow instructions
-from the GFPGAN's repository until `inference_gfpgan.py` script works.
 
-The following files must exist:
+:: (optional) install GFPGAN to fix faces
+pip install git+https://github.com/TencentARC/GFPGAN.git
 
-- `/sd/GFPGAN/inference_gfpgan.py`
+:: go into stable diffusion's repo directory
-- `/sd/GFPGAN/experiments/pretrained_models/GFPGANv1.3.pth`
+cd stable-diffusion
 
-If the GFPGAN directory does not exist, you will not get the option to use GFPGAN in the UI. If it does exist, you will either be able
+:: clone web ui
-to use it, or there will be a message in console with an error related to GFPGAN.
+git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
 
-### Web UI
+:: install requirements of web ui
+pip install -r stable-diffusion-webui/requirements.txt
 
-Run the script as:
+:: (outside of command line) put stable diffusion model into models/ldm/stable-diffusion-v1/model.ckpt; you'll have
+:: to create one missing directory;
+:: the command below must output something like: 1 File(s) 4,265,380,512 bytes
+dir models\ldm\stable-diffusion-v1\model.ckpt
 
-`python webui.py`
+:: (outside of command line) put the GFPGAN model into same directory as webui script
+:: the command below must output something like: 1 File(s) 348,632,874 bytes
+dir stable-diffusion-webui\GFPGANv1.3.pth
+```
+
+After that the installation is finished.
 
-When running the script, you must be in the main Stable Diffusion directory, `/sd`. If you cloned this repository into a subdirectory 
+Run the command to start web ui:
-of `/sd`, say, the `stable-diffusion-webui` directory, you will run it as:
 
-`python stable-diffusion-webui/webui.py`
+```
+python stable-diffusion-webui/webui.py
+```
+
+If you have a 4GB video card, run the command with `--lowvram` argument:
+
+```
+python stable-diffusion-webui/webui.py --lowvram
+```
 
-When launching, you may get a very long warning message related to some weights not being used. You may freely ignore it.
 After a while, you will get a message like this:
 
 ```

requirements.txt

@@ -0,0 +1,10 @@
+basicsr
+gfpgan
+gradio
+numpy
+Pillow
+realesrgan
+torch
+transformers
+omegaconf
+pytorch_lightning

webui.py

@@ -1,8 +1,18 @@
 import argparse
 import os
 import sys
-from collections import namedtuple
+
-from contextlib import nullcontext
+script_path = os.path.dirname(os.path.realpath(__file__))
+sd_path = os.path.dirname(script_path)
+
+# add parent directory to path; this is where Stable diffusion repo should be
+path_dirs = [(sd_path, 'ldm', 'Stable Diffusion'), ('../../taming-transformers', 'taming', 'Taming Transformers')]
+for d, must_exist, what in path_dirs:
+    must_exist_path = os.path.abspath(os.path.join(script_path, d, must_exist))
+    if not os.path.exists(must_exist_path):
+        print(f"Warning: {what} not found at path {must_exist_path}", file=sys.stderr)
+    else:
+        sys.path.append(os.path.join(script_path, d))
 
 import torch
 import torch.nn as nn
@@ -19,6 +29,9 @@ import html
 import time
 import json
 import traceback
+from collections import namedtuple
+from contextlib import nullcontext
+import signal
 
 import k_diffusion.sampling
 from ldm.util import instantiate_from_config
@@ -33,7 +46,6 @@ gradio.utils.get_local_ip_address = lambda: '127.0.0.1'
 mimetypes.init()
 mimetypes.add_type('application/javascript', '.js')
 
-script_path = os.path.dirname(os.path.realpath(__file__))
 
 # 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
@@ -44,9 +56,10 @@ invalid_filename_chars = '<>:"/\\|?*\n'
 config_filename = "config.json"
 
 parser = argparse.ArgumentParser()
-parser.add_argument("--config", type=str, default="configs/stable-diffusion/v1-inference.yaml", help="path to config which constructs model",)
+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("--ckpt", type=str, default="models/ldm/stable-diffusion-v1/model.ckpt", help="path to checkpoint of model",)
+parser.add_argument("--ckpt", type=str, default=os.path.join(sd_path, "models/ldm/stable-diffusion-v1/model.ckpt"), help="path to checkpoint of model",)
 parser.add_argument("--gfpgan-dir", type=str, help="GFPGAN directory", default=('./src/gfpgan' if os.path.exists('./src/gfpgan') else './GFPGAN'))
+parser.add_argument("--gfpgan-model", type=str, help="GFPGAN model file name", default='GFPGANv1.3.pth')
 parser.add_argument("--no-half", action='store_true', help="do not switch the model to 16-bit floats")
 parser.add_argument("--no-progressbar-hiding", action='store_true', help="do not hide progressbar in gradio UI (we hide it because it slows down ML if you have hardware accleration in browser)")
 parser.add_argument("--max-batch-count", type=int, default=16, help="maximum batch count value for the UI")
@@ -122,25 +135,34 @@ sd_upscalers = {
 }
 
 
-have_gfpgan = False
+def gfpgan_model_path():
-if os.path.exists(cmd_opts.gfpgan_dir):
+    places = [script_path, '.', os.path.join(cmd_opts.gfpgan_dir, 'experiments/pretrained_models')]
-    try:
+    files = [cmd_opts.gfpgan_model] + [os.path.join(dirname, cmd_opts.gfpgan_model) for dirname in places]
-        sys.path.append(os.path.abspath(cmd_opts.gfpgan_dir))
+    found = [x for x in files if os.path.exists(x)]
-        from gfpgan import GFPGANer
+
+    if len(found) == 0:
+        raise Exception("GFPGAN model not found in paths: " + ", ".join(files))
 
-        have_gfpgan = True
+    return found[0]
-    except:
-        print("Error importing GFPGAN:", file=sys.stderr)
-        print(traceback.format_exc(), file=sys.stderr)
 
 
 def gfpgan():
-    model_name = 'GFPGANv1.3'
+    return GFPGANer(model_path=gfpgan_model_path(), upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None)
-    model_path = os.path.join(cmd_opts.gfpgan_dir, 'experiments/pretrained_models', model_name + '.pth')
+
-    if not os.path.isfile(model_path):
+
-        raise Exception("GFPGAN model not found at path "+model_path)
+have_gfpgan = False
+try:
+    model_path = gfpgan_model_path()
+
+    if os.path.exists(cmd_opts.gfpgan_dir):
+        sys.path.append(os.path.abspath(cmd_opts.gfpgan_dir))
+    from gfpgan import GFPGANer
+
+    have_gfpgan = True
+except Exception:
+    print("Error setting up GFPGAN:", file=sys.stderr)
+    print(traceback.format_exc(), file=sys.stderr)
 
-    return GFPGANer(model_path=model_path, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None)
 
 
 class Options:
@@ -865,6 +887,7 @@ def process_images(p: StableDiffusionProcessing) -> Processed:
     seed = int(random.randrange(4294967294) if p.seed == -1 else p.seed)
 
     sample_path = os.path.join(p.outpath, "samples")
+    os.makedirs(sample_path, exist_ok=True)
     base_count = len(os.listdir(sample_path))
     grid_count = len(os.listdir(p.outpath)) - 1
 
@@ -1669,5 +1692,12 @@ demo = gr.TabbedInterface(
     analytics_enabled=False,
 )
 
+# make the program just exit at ctrl+c without waiting for anything
+def sigint_handler(signal, frame):
+    print('Interrupted')
+    os._exit(0)
+
+signal.signal(signal.SIGINT, sigint_handler)
+
 demo.queue(concurrency_count=1)
 demo.launch()