OPENGIT/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-04-29 16:29:01 +08:00
Code Issues Packages Projects Releases Wiki Activity
stable-diffusion-webui/modules
History
Alexander Ljungberg d9cc0910c8
Fix upcast attention dtype error. 
Without this fix, enabling the "Upcast cross attention layer to float32" option while also using `--opt-sdp-attention` breaks generation with an error:

```
  File "/ext3/automatic1111/stable-diffusion-webui/modules/sd_hijack_optimizations.py", line 612, in sdp_attnblock_forward
    out = torch.nn.functional.scaled_dot_product_attention(q, k, v, dropout_p=0.0, is_causal=False)
RuntimeError: Expected query, key, and value to have the same dtype, but got query.dtype: float key.dtype: float and value.dtype: c10::Half instead.
```

The fix is to make sure to upcast the value tensor too.
2023-06-06 21:45:30 +01:00
..
api
add /sdapi/v1/script-info api
2023-05-17 22:43:24 +03:00
codeformer
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
hypernetworks
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
models/diffusion
Merge pull request #10268 from Sakura-Luna/pbar
2023-05-11 08:16:36 +03:00
textual_inversion
Merge pull request #10780 from akx/image-emb-fonts
2023-06-02 14:36:22 +03:00
call_queue.py
Fix up string formatting/concatenation to f-strings where feasible
2023-05-09 22:25:39 +03:00
cmd_args.py
revert default cross attention optimization to Doggettx
2023-06-01 08:12:21 +03:00
codeformer_model.py
fixes for B007
2023-05-10 11:37:18 +03:00
config_states.py
load extensions' git metadata in parallel to loading the main program to save a ton of time during startup
2023-05-15 20:57:11 +03:00
deepbooru_model.py
Add option for float32 sampling with float16 UNet
2023-01-25 01:13:02 -05:00
deepbooru.py
ruff auto fixes
2023-05-10 11:05:02 +03:00
devices.py
ruff auto fixes
2023-05-10 11:05:02 +03:00
errors.py
better support for xformers flash attention on older versions of torch
2023-01-23 16:40:20 +03:00
esrgan_model_arch.py
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
esrgan_model.py
change upscalers to download models into user-specified directory (from commandline args) rather than the default models/<...>
2023-05-19 09:09:18 +03:00
extensions.py
revert git describe --always --tags for extensions because it seems to be causing issues
2023-05-21 13:30:09 +03:00
extra_networks_hypernet.py
manual fixes for ruff
2023-05-10 08:25:25 +03:00
extra_networks.py
fix [Bug]: LoRA don't apply on dropdown list sd_lora #10880
2023-05-31 22:57:27 +03:00
extras.py
use a local variable instead of dictionary entry for sd_merge_models in merge model metadata code
2023-05-17 17:44:07 +03:00
face_restoration.py
codeformer support
2022-09-07 12:32:28 +03:00
generation_parameters_copypaste.py
add quoting for infotext values that have a colon in them
2023-05-31 22:54:00 +03:00
gfpgan_model.py
F401 fixes for ruff
2023-05-10 09:02:23 +03:00
hashes.py
calculate hashes for Lora
2023-05-19 22:59:29 +03:00
images.py
Merge pull request #10905 from AUTOMATIC1111/fix-10896-pnginfo-parameters
2023-06-02 10:37:35 +03:00
img2img.py
imports cleanup for ruff
2023-05-10 08:43:42 +03:00
import_hook.py
add a comment for disable xformers hack
2022-12-24 07:57:56 +03:00
interrogate.py
fixes for B007
2023-05-10 11:37:18 +03:00
launch_utils.py
reworking launch.py: add references to renamed file
2023-05-21 16:27:34 +03:00
localization.py
Merge pull request #10335 from akx/l10n-dis-take-2
2023-05-13 20:46:50 +03:00
lowvram.py
Add support for the Variations models (unclip-h and unclip-l)
2023-03-25 21:03:07 -04:00
mac_specific.py
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
masking.py
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
memmon.py
attempt to fix memory monitor with multiple CUDA devices
2023-03-12 21:04:17 +03:00
modelloader.py
change upscalers to download models into user-specified directory (from commandline args) rather than the default models/<...>
2023-05-19 09:09:18 +03:00
ngrok.py
Merge branch 'dev' into ngrok-py
2023-05-18 10:12:17 +03:00
paths_internal.py
fix --data-dir for COMMANDLINE_ARGS
2023-05-12 22:33:21 +09:00
paths.py
F401 fixes for ruff
2023-05-10 09:02:23 +03:00
postprocessing.py
Merge branch 'dev' into master
2023-04-29 17:02:14 +03:00
processing.py
Merge pull request #10804 from AUTOMATIC1111/fix-xyz-clip
2023-05-31 22:54:51 +03:00
progress.py
move some settings to the new Optimization page
2023-05-17 20:22:54 +03:00
prompt_parser.py
fixes for B007
2023-05-10 11:37:18 +03:00
realesrgan_model.py
change upscalers to download models into user-specified directory (from commandline args) rather than the default models/<...>
2023-05-19 09:09:18 +03:00
Roboto-Regular.ttf
Drop fonts + font-roboto deps since we only use the single regular cut of Roboto
2023-05-11 10:07:28 +03:00
safe.py
Merge pull request #10339 from catboxanon/bf16
2023-05-13 20:21:39 +03:00
script_callbacks.py
make it possible for scripts to add cross attention optimizations
2023-05-18 22:48:28 +03:00
script_loading.py
F401 fixes for ruff
2023-05-10 09:02:23 +03:00
scripts_auto_postprocessing.py
ruff auto fixes
2023-05-10 11:05:02 +03:00
scripts_postprocessing.py
suggestions and fixes from the PR
2023-05-10 21:21:32 +03:00
scripts.py
Make load_scripts create new runners (removes reload_scripts)
2023-05-19 15:49:53 +03:00
sd_disable_initialization.py
autofixes from ruff
2023-05-10 07:52:45 +03:00
sd_hijack_checkpoint.py
add option to show/hide warnings
2023-01-18 23:04:24 +03:00
sd_hijack_clip_old.py
Fix up string formatting/concatenation to f-strings where feasible
2023-05-09 22:25:39 +03:00
sd_hijack_clip.py
suggestions and fixes from the PR
2023-05-10 21:21:32 +03:00
sd_hijack_inpainting.py
manual fixes for some C408
2023-05-10 11:55:09 +03:00
sd_hijack_ip2p.py
imports cleanup for ruff
2023-05-10 08:43:42 +03:00
sd_hijack_open_clip.py
Add support Stable Diffusion 2.0
2022-11-26 16:10:46 +03:00
sd_hijack_optimizations.py
Fix upcast attention dtype error.
2023-06-06 21:45:30 +01:00
sd_hijack_unet.py
Fix up string formatting/concatenation to f-strings where feasible
2023-05-09 22:25:39 +03:00
sd_hijack_utils.py
Fix full previews, --no-half-vae
2023-01-26 01:43:35 -05:00
sd_hijack_xlmr.py
imports cleanup for ruff
2023-05-10 08:43:42 +03:00
sd_hijack.py
revert default cross attention optimization to Doggettx
2023-06-01 08:12:21 +03:00
sd_models_config.py
imports cleanup for ruff
2023-05-10 08:43:42 +03:00
sd_models.py
assign devices.dtype early because it's needed before the model is loaded
2023-06-01 07:28:46 +03:00
sd_samplers_common.py
simplify single_sample_to_image
2023-05-17 14:53:39 +03:00
sd_samplers_compvis.py
suppress ENSD infotext for samplers that don't use it
2023-05-16 11:54:02 +03:00
sd_samplers_kdiffusion.py
Discard penultimate sigma for dpmpp_2m_sde
2023-05-21 01:01:59 -04:00
sd_samplers.py
suppress ENSD infotext for samplers that don't use it
2023-05-16 11:54:02 +03:00
sd_vae_approx.py
modules/sd_vae_approx.py: fix VAE-approx path
2023-03-11 20:00:12 +08:00
sd_vae_taesd.py
TAESD fix
2023-05-17 17:39:07 +08:00
sd_vae.py
Merge branch 'dev' into find_vae
2023-05-14 11:46:27 +03:00
shared_items.py
make it possible for scripts to add cross attention optimizations
2023-05-18 22:48:28 +03:00
shared.py
assign devices.dtype early because it's needed before the model is loaded
2023-06-01 07:28:46 +03:00
styles.py
alternative solution to fix styles load when edited by human #9765 as suggested by akx
2023-05-17 22:50:08 +03:00
sub_quadratic_attention.py
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00
timer.py
display correct timings after restarting UI
2023-03-12 21:25:22 +03:00
txt2img.py
rework hires prompts/sampler code to among other things support different extra networks in first/second pass
2023-05-18 20:16:09 +03:00
ui_common.py
bump gradio to 3.27
2023-04-29 09:17:35 +03:00
ui_components.py
Allow user input for gradio theme selection
2023-04-29 12:52:09 +03:00
ui_extensions.py
--filter=blob:none
2023-05-21 16:42:54 +09:00
ui_extra_networks_checkpoints.py
save previews for extra networks in the selected format
2023-03-11 15:46:20 +03:00
ui_extra_networks_hypernets.py
save previews for extra networks in the selected format
2023-03-11 15:46:20 +03:00
ui_extra_networks_textual_inversion.py
save previews for extra networks in the selected format
2023-03-11 15:46:20 +03:00
ui_extra_networks.py
Deduplicate default extra network registration
2023-05-19 15:35:16 +03:00
ui_loadsave.py
allow hiding buttons in ui-config.json
2023-05-19 19:04:20 +03:00
ui_postprocessing.py
imports cleanup for ruff
2023-05-10 08:43:42 +03:00
ui_tempdir.py
fix linter issue for 1.3.0
2023-05-27 20:23:16 +03:00
ui.py
Merge pull request #10785 from nyqui/fix-hires.fix
2023-05-31 22:54:24 +03:00
upscaler.py
change upscalers to download models into user-specified directory (from commandline args) rather than the default models/<...>
2023-05-19 09:09:18 +03:00
xlmr.py
Autofix Ruff W (not W605) (mostly whitespace)
2023-05-11 20:29:11 +03:00