Lee Bousfield
f9706acf43
Support loading textual inversion embeddings from safetensors files
2023-01-10 18:40:34 -07:00
AUTOMATIC
29fb532764
change color selector in settings to be part of form
2023-01-10 23:47:07 +03:00
AUTOMATIC
0f8603a559
add support for transformers==4.25.1
...
add fallback for when quick model creation fails
2023-01-10 17:46:59 +03:00
AUTOMATIC
ce3f639ec8
add more stuff to ignore when creating model from config
...
prevent .vae.safetensors files from being listed as stable diffusion models
2023-01-10 16:51:04 +03:00
AUTOMATIC
0c3feb202c
disable torch weight initialization and CLIP downloading/reading checkpoint to speedup creating sd model from config
2023-01-10 14:08:29 +03:00
AUTOMATIC
ef75c98053
Split history ui.py to ui_progress.py
2023-01-10 12:29:45 +03:00
Andrey
54dd5d6634
Split history ui.py to ui_progress.py
2023-01-10 11:54:49 +03:00
Andrey
f9c2147dfb
Split history ui.py to ui_progress.py
2023-01-10 11:54:49 +03:00
Andrey
27ea6949d3
Split history ui.py to ui_progress.py
2023-01-10 11:54:48 +03:00
Andrey
e9f8292a3a
Split history ui.py to ui_progress.py
2023-01-10 11:54:48 +03:00
aria1th
a4a5475cfa
Variable dropout rate
...
Implements variable dropout rate from #4549
Fixes hypernetwork multiplier being able to modified during training, also fixes user-errors by setting multiplier value to lower values for training.
Changes function name to match torch.nn.module standard
Fixes RNG reset issue when generating previews by restoring RNG state
2023-01-10 14:56:57 +09:00
AUTOMATIC1111
bd4587d2f5
Merge pull request #6578 from vladmandic/fix-model-load
...
allow model load if previous model failed
2023-01-10 08:22:44 +03:00
Vladimir Mandic
2275f130bf
relax reponse type check enforcement
2023-01-09 21:23:58 -05:00
Vladimir Mandic
552d7b90bf
allow model load if previous model failed
2023-01-09 18:34:26 -05:00
AUTOMATIC
3fe9e9e54d
fix broken resolution detection when pasting parameters with old hires fix enabled
2023-01-10 02:17:42 +03:00
AUTOMATIC1111
b1d976dca2
Merge pull request #6466 from vladmandic/api-get-memory
...
Implement API get-memory
2023-01-10 02:02:19 +03:00
Vladimir Mandic
95727312ca
remove bytes -> gb conversion
2023-01-09 16:54:12 -05:00
AUTOMATIC
1fbb6f9ebe
make a dropdown for prompt template selection
2023-01-09 23:35:40 +03:00
AUTOMATIC
43bb5190fc
remove/simplify some changes from #6481
2023-01-09 22:52:23 +03:00
AUTOMATIC1111
18c001792a
Merge branch 'master' into varsize
2023-01-09 22:45:39 +03:00
AUTOMATIC
cdfcbd9959
Remove fallback for Protocol import and remove Protocol import and remove instances of Protocol in code
...
add some whitespace between functions to be in line with other code in the repo
2023-01-09 20:08:48 +03:00
AUTOMATIC1111
89c3663080
Merge pull request #6482 from ProGamerGov/patch-6
...
Add fallback for Protocol import
2023-01-09 20:06:50 +03:00
AUTOMATIC
49c4509ce2
use existing function for loading VAE weights from file
2023-01-09 19:58:35 +03:00
AUTOMATIC1111
99da2c5af6
Merge pull request #6528 from PlasmaPower/vae-safetensors
...
Add support for loading VAEs from safetensors files
2023-01-09 19:40:28 +03:00
AUTOMATIC
d4fd2418ef
add an option to use old hiresfix width/height behavior
...
add a visual effect to inactive hires fix elements
2023-01-09 14:57:47 +03:00
Lee Bousfield
cb255faec6
Add support for loading VAEs from safetensor files
2023-01-08 10:17:50 -07:00
AUTOMATIC1111
0d194e4ecc
Merge pull request #6465 from brkirch/fix-training
...
Fix training for newer PyTorch builds
2023-01-08 16:16:31 +03:00
AUTOMATIC
137ce534b2
remove some code duplication
...
remove calls to locals()
add a test for img2img with script
2023-01-08 16:14:38 +03:00
AUTOMATIC1111
e7f2f1e1b6
Merge pull request #6469 from noodleanon/scripts-from-api
...
Run scripts from API
2023-01-08 15:47:19 +03:00
noodleanon
6d0cc1e239
Corrected is_img2img param
2023-01-08 11:03:48 +00:00
AUTOMATIC
085427de0e
make it possible for extensions/scripts to add their own embedding directories
2023-01-08 09:37:33 +03:00
AUTOMATIC
a0c87f1fdf
skip images in embeddings dir if they have a second .preview extension
2023-01-08 08:52:26 +03:00
ProGamerGov
984b86dd0a
Add fallback for Protocol import
2023-01-07 13:08:21 -07:00
dan
72497895b9
Move batchsize check
2023-01-08 02:57:36 +08:00
dan
669fb18d52
Add checkbox for variable training dims
2023-01-08 02:31:40 +08:00
dan
448b9cedab
Allow variable img size
2023-01-08 02:14:36 +08:00
noodleanon
d38ede71d5
Added script support in txt2img endpoint
2023-01-07 14:21:31 +00:00
noodleanon
50e2536279
Merge branch 'AUTOMATIC1111:master' into img2img-api-scripts
2023-01-07 14:18:09 +00:00
Vladimir Mandic
47534577ed
api-get-memory
2023-01-07 07:51:35 -05:00
brkirch
df3b31eb55
In-place operations can break gradient calculation
2023-01-07 07:04:59 -05:00
AUTOMATIC
fdfce47110
add "from" resolution for hires fix to be less confusing.
2023-01-07 13:29:47 +03:00
AUTOMATIC1111
983167e621
Merge pull request #6448 from aednzxy/patch-2
...
increase upscale api validation limit
2023-01-07 12:34:42 +03:00
AUTOMATIC1111
c295e4a244
Merge pull request #6055 from brkirch/sub-quad_attn_opt
...
Add Birch-san's sub-quadratic attention implementation
2023-01-07 12:26:55 +03:00
AUTOMATIC
1a5b86ad65
rework hires fix preview for #6437 : movie it to where it takes less place, make it actually account for all relevant sliders and calculate dimensions correctly
2023-01-07 09:56:37 +03:00
AUTOMATIC
de97380445
this breaks on default config because width, height, hr_scale are None at that point.
2023-01-07 08:53:53 +03:00
AUTOMATIC1111
01cc07b81a
Merge pull request #6437 from Mitchell1711/show-target-resolution
...
Show upscaled resolution on hires fix
2023-01-07 08:43:28 +03:00
AUTOMATIC
c4a221c405
Merge branch 'clip_hijack_rework'
2023-01-07 08:43:08 +03:00
AUTOMATIC
1740c33547
more comments
2023-01-07 07:48:44 +03:00
AUTOMATIC
08066676a4
make it not break on empty inputs; thank you tarded, we are
2023-01-07 07:22:07 +03:00
Mitchell Boot
f94cfc563b
Changed HTML to textbox instead
...
Using HTML caused an issue where the row would expand for a frame when changing the sliders because of the loading animation. This solution also doesn't use any additional HTML padding
2023-01-07 01:15:22 +01:00