cc8c9b7474
fix broken calls to find_checkpoint_config
3 years ago
9beb794e0b
clarify the option to disable NaN check.
3 years ago
6f31d2210c
support detecting midas model
...
fix broken api for checkpoint list
3 years ago
d2ac95fa7b
remove the need to place configs near models
3 years ago
7a14c8ab45
add an option to enable sections from extras tab in txt2img/img2img
...
fix some style inconsistenices
3 years ago
10421f93c3
Fix full previews, --no-half-vae
3 years ago
6cff440182
fix prompt editing break after first batch in img2img
3 years ago
d1d6ce2983
add edit_image_conditioning from my earlier edits in case there's an attempt to inegrate pix2pix properly
...
this allows to use pix2pix model in img2img though it won't work well this way
3 years ago
3cead6983e
Merge pull request #7197 from mcmonkey4eva/fix-ti-symlinks
...
allow symlinks in the textual inversion embeddings folder
3 years ago
15e89ef0f6
fix for unet hijack breaking the train tab
3 years ago
789d47f832
make clicking extra networks button one more time close the extra networks UI
3 years ago
e179b6098a
allow symlinks in the textual inversion embeddings folder
3 years ago
1574e96729
Merge pull request #6510 from brkirch/unet16-upcast-precision
...
Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half
3 years ago
57c1baa774
change to code for live preview fix on OSX to be bit more obvious
3 years ago
23dafe6d86
Merge pull request #7151 from brkirch/fix-approx-nn
...
Fix Approx NN previews changing first generation result
3 years ago
ee0a0da324
Add instruct-pix2pix hijack
...
Allows loading instruct-pix2pix models via same method as inpainting models in sd_models.py and sd_hijack_ip2p.py
Adds ddpm_edit.py necessary for instruct-pix2pix
3 years ago
d5ce044bcd
Merge pull request #7146 from EllangoK/master
...
Adds X/Y/Z Grid Script
3 years ago
e3b53fd295
Add UI setting for upcasting attention to float32
...
Adds "Upcast cross attention layer to float32" option in Stable Diffusion settings. This allows for generating images using SD 2.1 models without --no-half or xFormers.
In order to make upcasting cross attention layer optimizations possible it is necessary to indent several sections of code in sd_hijack_optimizations.py so that a context manager can be used to disable autocast. Also, even though Stable Diffusion (and Diffusers) only upcast q and k, unfortunately my findings were that most of the cross attention layer optimizations could not function unless v is upcast also.
3 years ago
84d9ce30cb
Add option for float32 sampling with float16 UNet
...
This also handles type casting so that ROCm and MPS torch devices work correctly without --no-half. One cast is required for deepbooru in deepbooru_model.py, some explicit casting is required for img2img and inpainting. depth_model can't be converted to float16 or it won't work correctly on some systems (it's known to have issues on MPS) so in sd_models.py model.depth_model is removed for model.half().
3 years ago
5228ec8bda
remove fairscale requirement, add fake fairscale to make BLIP not complain about it mk2
3 years ago
28189985e6
remove fairscale requirement, add fake fairscale to make BLIP not complain about it
3 years ago
e46bfa5a9e
handling sub grids and merging into one
3 years ago
602a1864b0
also return the removed field to sdapi/v1/upscalers because someone might have relied on it existing
3 years ago
42a70d7477
repair sdapi/v1/upscalers returning bogus results
3 years ago
f64af77adc
Fix different first gen with Approx NN previews
...
The loading of the model for approx nn live previews can change the internal state of PyTorch, resulting in a different image. This can be avoided by preloading the approx nn model in advance.
3 years ago
45e270dfc8
add image decod exception handling
3 years ago
5c1cb9263f
fix BLIP failing to import depending on configuration
3 years ago
7ba7f4ed6e
Merge pull request #7113 from vladmandic/interrogate
...
Add selector to interrogate categories
3 years ago
7b1c7ba87b
add support for apostrophe in extra network names
3 years ago
04a561c11c
add option to skip interrogate categories
3 years ago
efa7287be0
Merge branch 'AUTOMATIC1111:master' into interrogate
3 years ago
171a5b3bb9
Merge pull request #7032 from gmq/extra-network-styles
...
Extra network view style
3 years ago
6e1b296baf
api-image-format
3 years ago
e8c3d03f7d
a possible fix for broken image upscaling
3 years ago
925dd09c91
improve interrogate
3 years ago
59146621e2
better support for xformers flash attention on older versions of torch
3 years ago
3fa482076a
Merge remote-tracking branch 'takuma104/xformers-flash-attention'
3 years ago
194cbd065e
fix open directory button failing
3 years ago
663353098e
Merge pull request #7031 from EllangoK/master
...
Fixes various button overflowing UI and compact checkbox
3 years ago
74608300d1
Merge pull request #7093 from Shondoit/fix-dark-mode
...
Fix dark mode
3 years ago
41265a026d
third time's the charm
3 years ago
fabdae089e
add missing import to previous commit
3 years ago
669dbd9725
Fix dark mode
...
Fixes #7048
Co-Authored-By: J.J. Tolton <jjtolton@gmail.com>
3 years ago
b5230197a6
rework extras tab to use script system
3 years ago
f80ff3c1e4
feat(extra-networks): remove view dropdown
3 years ago
66eef11ce7
feat(extra-networks): add default view setting
3 years ago
985c0b8e9a
feat(extra-networks): add thumbs view style
3 years ago
68303c96e5
split oversize extras.py to postprocessing.py
3 years ago
c56b367122
Split history extras.py to postprocessing.py
3 years ago
d63340a485
Split history extras.py to postprocessing.py
3 years ago
AUTOMATIC
brkirch
Alex "mcmonkey" Goodwin
Kyle
EllangoK
Vladimir Mandic
Shondoit
Guillermo Moreno