LLM/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-02-21 10:20:35 +08:00
Code Issues Packages Projects Releases Wiki Activity
4,913 Commits 749 Branches 94 Tags
4c5082ca685d1dd0a24907890da170e258e5cd01
Commit Graph

4913 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
M. Tolga Cangöz
c6fa2ee726 [Tests] Update a deprecated parameter in test files and fix several typos (#7277) 
* Add properties and `IPAdapterTesterMixin` tests for `StableDiffusionPanoramaPipeline`

* Fix variable name typo and update comments

* Update deprecated `output_type="numpy"` to "np" in test files

* Discard changes to src/diffusers/pipelines/stable_diffusion_panorama/pipeline_stable_diffusion_panorama.py

* Update test_stable_diffusion_panorama.py

* Update numbers in README.md

* Update get_guidance_scale_embedding method to use timesteps instead of w

* Update number of checkpoints in README.md

* Add type hints and fix var name

* Fix PyTorch's convention for inplace functions

* Fix a typo

* Revert "Fix PyTorch's convention for inplace functions"

This reverts commit 74350cf65b.

* Fix typos

* Indent

* Refactor get_guidance_scale_embedding method in LEditsPPPipelineStableDiffusionXL class
 2024-12-23 13:02:09 +05:30
Dhruv Nair
65eb818a38 Update Cascade Tests (#7324) 
* update

* update

* update
 2024-12-23 13:02:09 +05:30
Linoy Tsaban
f86aa42f4c [Advanced DreamBooth LoRA SDXL] Support EDM-style training (follow up of #7126) (#7182) 
* add edm style training

* style

* finish adding edm training feature

* import fix

* fix latents mean

* minor adjustments

* add edm to readme

* style

* fix autocast and scheduler config issues when using edm

* style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Suraj Patil
c7c339555c add edm schedulers in doc (#7319) 
* add edm schedulers in doc

* add in toctree

* address reviewe comments
 2024-12-23 13:02:09 +05:30
Anatoly Belikov
ec2acc7dc8 log loss per image (#7278) 
* log loss per image

* add commandline param for per image loss logging

* style

* debug-loss -> debug_loss

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Dhruv Nair
c159e36f61 [Tests] Fix incorrect constant in VAE scaling test. (#7301) 
update
 2024-12-23 13:02:09 +05:30
Sayak Paul
14ddd9612a add: pytest log installation (#7313) 2024-12-23 13:02:09 +05:30
Kenneth Gerald Hamilton
94c0940aef update get_order_list if statement (#7309) 
* update get_order_list if statement

* revery
 2024-12-23 13:02:09 +05:30
Beinsezii
e6851daf70 Change step_offset scheduler docstrings (#7128) 
* Change step_offset scheduler docstrings

* Mention it may be needed by some models

* More docstrings

These ones failed literal S&R because I performed it case-sensitive
which is fun.

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Alexander Bonnet
58de563cec Fix typos in UNet2DConditionModel documentation (#7291) 
* fix typo in UNet2DConditionModel documentation

* Fix indentation that may fix doc rendering

* Fix squished doc lines
 2024-12-23 13:02:09 +05:30
Michael
f08278b391 Add Intro page of TCD (#7259) 
* add tcd intro

* resolve repos

* Apply suggestions from code review

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* revise NFEs related

* change inpainting location

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-12-23 13:02:09 +05:30
jnhuang
e6e832c0a1 Fix Wrong Text-encoder Grad Setting in Custom_Diffusion Training (#7302) 
fix index in set textencoder grad

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
1cd2104a7e add: support for notifying maintainers about the nightly test status (#7117) 
* add: support for notifying maintainers about the nightly test status

* add: a tempoerary workflow for validation.

* cancel in progress.

* runs-on

* clean up

* add: peft dep

* change device.

* multiple edits.

* remove temp workflow.
 2024-12-23 13:02:09 +05:30
Sayak Paul
59aa7e0fb2 [PyPI publishing] feat: automate the process of pypi publication to some extent. (#7270) 
* feat: automate the process of pypi publication to some extent.

* utility to fetch the latest release branch

* correct package name.
 2024-12-23 13:02:09 +05:30
Manuel Brack
87c83dd381 [Pipeline] Add LEDITS++ pipelines (#6074) 
* Setup LEdits++ file structure

* Fix import

* LEditsPP Stable Diffusion pipeline

* Include variable image aspect ratios

* Implement LEDITS++ for SDXL

* clean up LEditsPPPipelineStableDiffusion

* Adjust inversion output

* Added docu, more cleanup for LEditsPPPipelineStableDiffusion

* clean up LEditsPPPipelineStableDiffusionXL

* Update documentation

* Fix documentation import

* Add skeleton IF implementation

* Fix documentation typo

* Add LEDTIS docu to toctree

* Add missing title

* Finalize SD documentation

* Finalize SD-XL documentation

* Fix code style and quality

* Fix typo

* Fix return types

* added LEditsPPPipelineIF; minor changes for LEditsPPPipelineStableDiffusion and LEditsPPPipelineStableDiffusionXL

* Fix copy reference

* add documentation for IF

* Add first tests

* Fix batching for SD-XL

* Fix text encoding and perfect reconstruction for SD-XL

* Add tests for SD-XL, minor changes

* move user_mask to correct device, use cross_attention_kwargs also for inversion

* Example docstring

* Fix attention resolution for non-square images

* Refactoring for PR review

* Safely remove ledits_utils.py

* Style fixes

* Replace assertions with ValueError

* Remove LEditsPPPipelineIF

* Remove unecessary input checks

* Refactoring of CrossAttnProcessor

* Revert unecessary changes to scheduler

* Remove first progress-bar in inversion

* Refactor scheduler usage and reset

* Use imageprocessor instead of custom logic

* Fix scheduler init warning

* Fix error when running the pipeline in fp16

* Update documentation wrt perfect inversion

* Update tests

* Fix code quality and copy consistency

* Update LEditsPP import

* Remove enable/disable methods that are now in StableDiffusionMixin

* Change import in docs

* Revert import structure change

* Fix ledits imports

---------

Co-authored-by: Katharina Kornmeier <katharina.kornmeier@stud.tu-darmstadt.de>
 2024-12-23 13:02:09 +05:30
Dhruv Nair
02672b9cc4 Update Stable Cascade Conversion Scripts (#7271) 
* update

* update

* update

* update

* update

* update

* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Dhruv Nair
8a422df965 Update Cascade documentation (#7257) 
* updates

* update

* update

* Update docs/source/en/api/pipelines/stable_cascade.md

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* update

* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Kashif Rasul <kashif.rasul@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
808cb88f28 [Chore] clean residue from copy-pasting in the UNet single file loader (#7295) 
clean residue from copy-pasting
 2024-12-23 13:02:09 +05:30
Dhruv Nair
a57a858d45 Fix loading Img2Img refiner components in from_single_file (#7282) 
* update

* update

* update

* update
 2024-12-23 13:02:09 +05:30
Dhruv Nair
3990d0a1ac Fix passing pooled prompt embeds to Cascade Decoder and Combined Pipeline (#7287) 
* update

* update

* update

* update
 2024-12-23 13:02:09 +05:30
Dhruv Nair
353856a185 Add single file support for Stable Cascade (#7274) 
* update

* update

* update

* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
8e6902da84 [LoRA] use the PyTorch classes wherever needed and start depcrecation cycles (#7204) 
* fix PyTorch classes and start deprecsation cycles.

* remove args crafting for accommodating scale.

* remove scale check in feedforward.

* assert against nn.Linear and not CompatibleLinear.

* remove conv_cls and lineaR_cls.

* remove scale

* 👋 scale.

* fix: unet2dcondition

* fix attention.py

* fix: attention.py again

* fix: unet_2d_blocks.

* fix-copies.

* more fixes.

* fix: resnet.py

* more fixes

* fix i2vgenxl unet.

* depcrecate scale gently.

* fix-copies

* Apply suggestions from code review

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* quality

* throw warning when scale is passed to the the BasicTransformerBlock class.

* remove scale from signature.

* cross_attention_kwargs, very nice catch by Yiyi

* fix: logger.warn

* make deprecation message clearer.

* address final comments.

* maintain same depcrecation message and also add it to activations.

* address yiyi

* fix copies

* Apply suggestions from code review

Co-authored-by: YiYi Xu <yixu310@gmail.com>

* more depcrecation

* fix-copies

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
dc7e9809a9 [Chore] switch to logger.warning (#7289) 
switch to logger.warning
 2024-12-23 13:02:09 +05:30
Dhruv Nair
a3768d47e0 Change export_to_video default (#6990) 
update

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
erliding
137e766b57 instruct pix2pix pipeline: remove sigma scaling when computing classifier free guidance (#7006) 
remove sigma scaling when computing cfg

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
fbc7065a76 [Dockerfiles] add: a workflow to check if docker containers can be built in case of modifications (#7129) 
* add: a workflow to check if docker containers can be built if the files are modified.

* type

* unify docker image build test and push

* make it run on prs too.

* check

* check

* check

* check again.

* remove docker test build file.

* remove extra dependencies./

* check
 2024-12-23 13:02:09 +05:30
pravdomil
87ce979fb6 Inline InputPadder (#6582) 
inline InputPadder
 2024-12-23 13:02:09 +05:30
Haofan Wang
77c6d6b44f Support latents_mean and latents_std (#7132) 
* update latents_mean and latents_std

* fix typos

* Update src/diffusers/pipelines/controlnet/pipeline_controlnet_sd_xl_img2img.py

* format

---------

Co-authored-by: ResearcherXman <xhs.research@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-12-23 13:02:09 +05:30
pravdomil
1abab1d268 Fix gmflow_dir (#6583) 
* remove sys.path

* update readme
 2024-12-23 13:02:09 +05:30
Sayak Paul
0305f84822 [Chore] remove tf mention (#7245) 
remove tf mention
 2024-12-23 13:02:09 +05:30
Aryan
77961cefcb [docs/nits] Fix return values based on return_dict and minor doc updates (#7105) 
* fix returns and docs

* handle latent output_type correctly

* revert to old tensor2vid impl

* make fix-copies

* fix return in community animatediff pipes

* fix return docstring

* fix return docs

* add missing quote

---------

Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
 2024-12-23 13:02:09 +05:30
Xiaodong Wang
701b172ead [UNet_Spatio_Temporal_Condition] fix default num_attention_heads in unet_spatio_temporal_condition (#7205) 
fix default num_attention_heads in unet_spatio_temporal_condition

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
YiYi Xu
d6e2444565 update the signature of from_single_file (#7216) 
* update the signature of from_single_file

* Update src/diffusers/loaders/single_file.py

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/loaders/single_file.py

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/loaders/single_file.py

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update src/diffusers/loaders/single_file.py

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-12-23 13:02:09 +05:30
qqii
b31c77fcb3 [Community Pipeline] Skip Marigold depth_colored with color_map=None (#7170) 
[Community Pipeline] Skip Marigold depth_colored generation by passing color_map=None
 2024-12-23 13:02:09 +05:30
Mengqing Cao
dccf392fe7 Add npu support (#7144) 
* Add npu support

* fix for code quality check

* fix for code quality check
 2024-12-23 13:02:09 +05:30
UmerHA
c5c113369b Adds denoising_end parameter to ControlNetPipeline for SDXL (#6175) 
* Initial commit

* Removed copy hints, as in original SDXLControlNetPipeline

Removed copy hints, as in original SDXLControlNetPipeline, as the `make fix-copies` seems to have issues with the @property decorator.

* Reverted changes to ControlNetXS

* Addendum to: Removed changes to ControlNetXS

* Added test+docs for mixture of denoiser

* Update docs/source/en/using-diffusers/controlnet.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/using-diffusers/controlnet.md

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-12-23 13:02:09 +05:30
Steven Liu
6b2f8109bc [docs] IP-Adapter image embedding (#7226) 
* update

* fix parameter name

* feedback

* add no mask version

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:09 +05:30
Sayak Paul
891c1d8031 [Tests] fix config checking tests (#7247) 
* debig

* cast tuples to lists.

* debug

* handle upcast attention

* handle downblock types for vae.

* remove print.

* address Dhruv's comments.

* fix: upblock types.

* upcast attention

* debug

* debug

* debug

* better guarding.

* style
 2024-12-23 13:02:09 +05:30
Chi
1197d7dca8 Solve missing clip_sample implementation in FlaxDDIMScheduler. (#7017) 
* I added a new doc string to the class. This is more flexible to understanding other developers what are doing and where it's using.

* Update src/diffusers/models/unet_2d_blocks.py

This changes suggest by maintener.

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update src/diffusers/models/unet_2d_blocks.py

Add suggested text

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update unet_2d_blocks.py

I changed the Parameter to Args text.

* Update unet_2d_blocks.py

proper indentation set in this file.

* Update unet_2d_blocks.py

a little bit of change in the act_fun argument line.

* I run the black command to reformat style in the code

* Update unet_2d_blocks.py

similar doc-string add to have in the original diffusion repository.

* Fix bug for mention in this issue section #6901

* Update src/diffusers/schedulers/scheduling_ddim_flax.py

Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* Fix linter

* Restore empty line

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
 2024-12-23 13:02:09 +05:30
Martin Müller
6f71a65f5b Make mid block optional for flax UNet (#7083) 
* make mid block optional for flax UNet

* make style
 2024-12-23 13:02:09 +05:30
Sayak Paul
2e3436d2dc [Easy] fix: save_model_card utility of the DreamBooth SDXL LoRA script (#7258) 
* fix: save_model_card utility.

* fix a little more to make it more lenient.

* remove lower()
 2024-12-23 13:02:08 +05:30
Sayak Paul
da58475f40 [Utils] Improve " # Copied from ..." statements in the pipelines (#6917) 
* copied from for t2i pipelines without ip adapter support.

* two more pipelines with proper copied from comments.

* revert to the original implementation
 2024-12-23 13:02:08 +05:30
Sayak Paul
b8c0fab3e8 [Tests] fix: VAE tiling tests when setting the right device (#7246) 
* debug

* checking

* fix more

* remove device.

* fix-copies
 2024-12-23 13:02:08 +05:30
Sayak Paul
b658f1f162 [Core] throw error when patch inputs and layernorm are provided for Transformers2D (#7200) 
* throw error when patch inputs and layernorm are provided for transformers2d.

* add comment on supported norm_types in transformers2d

* more check

* fix: norm _type handling
 2024-12-23 13:02:08 +05:30
Steven Liu
9d10e629af [docs] Community tips (#7137) 
* tips

* feedback

* callback only
 2024-12-23 13:02:08 +05:30
pravdomil
ffc485f3d9 use self.device (#6595) 
* use self.device

* use device

* fix

* fix
 2024-12-23 13:02:08 +05:30
Steven Liu
19d9e7d5d9 [docs] Merge LoRAs (#7213) 
* merge loras

* feedback

* torch.compile

* feedback
 2024-12-23 13:02:08 +05:30
bimsarapathiraja
e5b4915090 Remove the line. Using it create wrong output (#7075) 
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-12-23 13:02:08 +05:30
Dhruv Nair
3dd881488f Raise an error when trying to use SD Cascade Decoder with dtype bfloat16 and torch < 2.2 (#7244) 
update
 2024-12-23 13:02:08 +05:30
Sayak Paul
413c6e97f3 fix: support for loading playground v2.5 single file checkpoint. (#7230) 
* fix: support for loading playground v2.5 single file checkpoint.

* remove is_playground_model.

* fix: edm key

* apply Dhruv's comments but errors.

* fix: things.

* delegate model_type inference to a function.

* address Dhruv's comment.

* address rest of the comments.

* fix: kwargs

* fix

* update

---------

Co-authored-by: DN6 <dhruv.nair@gmail.com>
 2024-12-23 13:02:08 +05:30