AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00
Code Activity
4,601 Commits 740 Branches 94 Tags
31058cdaef63ca660a1a045281d156239fba8192
Commit Graph

4601 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sayak Paul
31058cdaef [LoRA] allow loras to be loaded with low_cpu_mem_usage. (#9510) 
* allow loras to be loaded with low_cpu_mem_usage.

* add flux support but note https://github.com/huggingface/diffusers/pull/9510\#issuecomment-2378316687

* low_cpu_mem_usage.

* fix-copies

* fix-copies again

* tests

* _LOW_CPU_MEM_USAGE_DEFAULT_LORA

* _peft_version default.

* version checks.

* version check.

* version check.

* version check.

* require peft 0.13.1.

* explicitly specify low_cpu_mem_usage=False.

* docs.

* transformers version 4.45.2.

* update

* fix

* empty

* better name initialize_dummy_state_dict.

* doc todos.

* Apply suggestions from code review

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

* style

* fix-copies

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-10-09 10:57:16 +05:30
Yijun Lee
ec9e5264c0 refac/pipeline_output (#9582) 2024-10-08 16:11:13 -10:00
sanaka
acd6d2c42f Fix the bug that joint_attention_kwargs is not passed to the FLUX's transformer attention processors (#9517) 
* Update transformer_flux.py
 2024-10-08 11:25:48 -10:00
v2ray
86bd991ee5 Fixed noise_pred_text referenced before assignment. (#9537) 
* Fixed local variable noise_pred_text referenced before assignment when using PAG with guidance scale and guidance rescale at the same time.

* Fixed style.

* Made returning text pred noise an argument.
 2024-10-08 09:27:10 -10:00
Sayak Paul
02eeb8e77e [LoRA] Handle DoRA better (#9547) 
* handle dora.

* print test

* debug

* fix

* fix-copies

* update logits

* add warning in the test.

* make is_dora check consistent.

* fix-copies
 2024-10-08 21:47:44 +05:30
glide-the
66eef9a6dc fix: CogVideox train dataset _preprocess_data crop video (#9574) 
* Removed int8 to float32 conversion (`* 2.0 - 1.0`) from `train_transforms` as it caused image overexposure.

Added `_resize_for_rectangle_crop` function to enable video cropping functionality. The cropping mode can be configured via `video_reshape_mode`, supporting options: ['center', 'random', 'none'].

* The number 127.5 may experience precision loss during division operations.

* wandb request pil image Type

* Resizing bug

* del jupyter

* make style

* Update examples/cogvideo/README.md

* make style

---------

Co-authored-by: --unset <--unset>
Co-authored-by: Aryan <aryan@huggingface.co>
 2024-10-08 12:52:52 +05:30
Sayak Paul
63a5c8742a Update distributed_inference.md to include transformer.device_map (#9553) 
* Update distributed_inference.md to include `transformer.device_map`

* Update docs/source/en/training/distributed_inference.md

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

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-10-08 08:03:51 +05:30
Eliseu Silva
1287822973 Fix for use_safetensors parameters, allow use of parameter on loading submodels (#9576) (#9587) 
* Fix for use_safetensors parameters, allow use of parameter on loading submodels (#9576)
 2024-10-07 10:41:32 -10:00
Yijun Lee
a80f689200 refac: docstrings in import_utils.py (#9583) 
* refac: docstrings in import_utils.py

* Update import_utils.py
 2024-10-07 13:27:35 -07:00
captainzz
2cb383f591 fix vae dtype when accelerate config using --mixed_precision="fp16" (#9601) 
* fix vae dtype when accelerate config using --mixed_precision="fp16"

* Add param for upcast vae
 2024-10-07 21:00:25 +05:30
Sayak Paul
31010ecc45 [Chore] add a note on the versions in Flux LoRA integration tests (#9598) 
add a note on the versions.
 2024-10-07 17:43:48 +05:30
Clem
3159e60d59 fix xlabs FLUX lora conversion typo (#9581) 
* fix startswith syntax in xlabs lora conversion

* Trigger CI

https://github.com/huggingface/diffusers/pull/9581#issuecomment-2395530360
 2024-10-07 10:47:54 +05:30
YiYi Xu
99f608218c [sd3] make sure height and size are divisible by 16 (#9573) 
* check size

* up
 2024-10-03 08:36:26 -10:00
Xiangchendong
7f323f0f31 fix cogvideox autoencoder decode (#9569) 
Co-authored-by: Aryan <aryan@huggingface.co>
 2024-10-02 09:07:06 -10:00
Darren Hsu
61d37640ad Support bfloat16 for Upsample2D (#9480) 
* Support bfloat16 for Upsample2D

* Add test and use is_torch_version

* Resolve comments and add decorator

* Simplify require_torch_version_greater_equal decorator

* Run make style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-10-01 16:08:12 -10:00
JuanCarlosPi
33fafe3d14 Add PAG support to StableDiffusionControlNetPAGInpaintPipeline (#8875) 
* Add pag to controlnet inpainting pipeline


---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-30 20:04:42 -10:00
hlky
c4a8979f30 Add beta sigmas to other schedulers and update docs (#9538) 2024-09-30 09:00:54 -10:00
Sayak Paul
f9fd511466 [LoRA] support Kohya Flux LoRAs that have text encoders as well (#9542) 
* support kohya flux loras that have tes.
 2024-09-30 07:59:39 -10:00
Sayak Paul
8e7d6c03a3 [chore] fix: retain memory utility. (#9543) 
* fix: retain memory utility.

* fix

* quality

* free_memory.
 2024-09-28 21:08:45 +05:30
Anand Kumar
b28675c605 [train_instruct_pix2pix.py]Fix the LR schedulers when num_train_epochs is passed in a distributed training env (#9316) 
Fixed pix2pix lr scheduler

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-09-28 21:01:37 +05:30
Aryan
bd4df2856a [refactor] remove conv_cache from CogVideoX VAE (#9524) 
* remove conv cache from the layer and pass as arg instead

* make style

* yiyi's cleaner implementation

Co-Authored-By: YiYi Xu <yixu310@gmail.com>

* sayak's compiled implementation

Co-Authored-By: Sayak Paul <spsayakpaul@gmail.com>

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-28 17:09:30 +05:30
Sayak Paul
11542431a5 [Core] fix variant-identification. (#9253) 
* fix variant-idenitification.

* fix variant

* fix sharded variant checkpoint loading.

* Apply suggestions from code review

* fixes.

* more fixes.

* remove print.

* fixes

* fixes

* comments

* fixes

* apply suggestions.

* hub_utils.py

* fix test

* updates

* fixes

* fixes

* Apply suggestions from code review

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

* updates.

* removep patch file.

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-28 09:57:31 +05:30
Sayak Paul
81cf3b2f15 [Tests] [LoRA] clean up the serialization stuff. (#9512) 
* clean up the serialization stuff.

* better
 2024-09-27 07:57:09 -10:00
PromeAI
534848c370 [examples] add train flux-controlnet scripts in example. (#9324) 
* add train flux-controlnet scripts in example.

* fix error

* fix subfolder error

* fix preprocess error

* Update examples/controlnet/README_flux.md

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

* Update examples/controlnet/README_flux.md

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

* fix readme

* fix note error

* add some Tutorial for deepspeed

* fix some Format Error

* add dataset_path example

* remove print, add guidance_scale CLI, readable apply

* Update examples/controlnet/README_flux.md

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

* update,push_to_hub,save_weight_dtype,static method,clear_objs_and_retain_memory,report_to=wandb

* add push to hub in readme

* apply weighting schemes

* add note

* Update examples/controlnet/README_flux.md

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

* make code style and quality

* fix some unnoticed error

* make code style and quality

* add example controlnet in readme

* add test controlnet

* rm Remove duplicate notes

* Fix formatting errors

* add new control image

* add model cpu offload

* update help for adafactor

* make quality & style

* make quality and style

* rename flux_controlnet_model_name_or_path

* fix back src/diffusers/pipelines/flux/pipeline_flux_controlnet.py

* fix dtype error by pre calculate text emb

* rm image save

* quality fix

* fix test

* fix tiny flux train error

* change report to to tensorboard

* fix save name error when test

* Fix shrinking errors

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Your Name <you@example.com>
 2024-09-27 13:31:47 +05:30
Sayak Paul
2daedc0ad3 [LoRA] make set_adapters() method more robust. (#9535) 
* make set_adapters() method more robust.

* remove patch

* better and concise code.

* Update src/diffusers/loaders/lora_base.py

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

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-27 07:32:43 +05:30
Aryan
665c6b47a2 [bug] Precedence of operations in VAE should be slicing -> tiling (#9342) 
* bugfix: precedence of operations should be slicing -> tiling

* fix typo

* fix another typo

* deprecate current implementation of tiled_encode and use new impl

* Update src/diffusers/models/autoencoders/autoencoder_kl.py

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

* Update src/diffusers/models/autoencoders/autoencoder_kl.py

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-26 22:12:07 +05:30
Álvaro Somoza
066ea374c8 [Tests] Fix ChatGLMTokenizer (#9536) 
fix
 2024-09-25 22:10:15 -10:00
YiYi Xu
9cd37557d5 flux controlnet fix (control_modes batch & others) (#9507) 
* flux controlnet mode to take into account batch size

* incorporate yiyixuxu's suggestions (cleaner logic) as well as clean up control mode handling for multi case

* fix

* fix use_guidance when controlnet is a multi and does not have config

---------

Co-authored-by: Christopher Beckham <christopher.j.beckham@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-09-25 19:09:54 -10:00
hlky
1c6ede9371 [Schedulers] Add beta sigmas / beta noise schedule (#9509) 
Add beta sigmas / beta noise schedule
 2024-09-25 13:30:32 -10:00
v2ray
aa3c46d99a [Doc] Improved level of clarity for latents_to_rgb. (#9529) 
Fixed latents_to_rgb doc.

Co-authored-by: Álvaro Somoza <asomoza@users.noreply.github.com>
 2024-09-25 19:26:58 -03:00
YiYi Xu
c76e88405c update get_parameter_dtype (#9526) 
* up

* Update src/diffusers/models/modeling_utils.py

Co-authored-by: Aryan <aryan@huggingface.co>

---------

Co-authored-by: Aryan <aryan@huggingface.co>
 2024-09-25 11:00:57 -10:00
Steven Liu
d9c969172d [docs] Model sharding (#9521) 
* flux shard

* feedback
 2024-09-25 09:33:54 -07:00
Lee Penkman
065ce07ac3 Update community_projects.md (#9266) 2024-09-25 08:54:36 -07:00
Sayak Paul
6ca5a58e43 [Community Pipeline] Batched implementation of Flux with CFG (#9513) 
* batched implementation of flux cfg.

* style.

* readme

* remove comments.
 2024-09-25 15:25:15 +05:30
hlky
b52684c3ed Add exponential sigmas to other schedulers and update docs (#9518) 2024-09-24 14:50:12 -10:00
YiYi Xu
bac8a2412d a few fix for SingleFile tests (#9522) 
* update sd15 repo

* update more
 2024-09-24 13:36:53 -10:00
Sayak Paul
28f9d84549 [CI] allow faster downloads from the Hub in CI. (#9478) 
* allow faster downloads from the Hub in CI.

* HF_HUB_ENABLE_HF_TRANSFER: 1

* empty

* empty

* remove ENV HF_HUB_ENABLE_HF_TRANSFER=1.

* empty
 2024-09-24 09:42:11 +05:30
LukeLin
2b5bc5be0b [Doc] Fix path and and also import imageio (#9506) 
* Fix bug

* import imageio
 2024-09-23 16:47:34 -07:00
captainzz
bab17789b5 fix bugs for sd3 controlnet training (#9489) 
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-23 13:40:44 -10:00
hlky
19547a5734 Add Noise Schedule/Schedule Type to Schedulers Overview documentation (#9504) 
* Add Noise Schedule/Schedule Type to Schedulers Overview docs

* Update docs/source/en/api/schedulers/overview.md

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

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
 2024-09-23 16:39:55 -07:00
Seongbin Lim
3e69e241f7 Allow DDPMPipeline half precision (#9222) 
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-23 13:28:14 -10:00
hlky
65f9439b56 [Schedulers] Add exponential sigmas / exponential noise schedule (#9499) 
* exponential sigmas

* Apply suggestions from code review

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

* make style

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-09-23 13:12:51 -10:00
pibbo88
00f5b41862 Fix the bug of sd3 controlnet training when using gradient checkpointing. (#9498) 
Fix the bug of sd3 controlnet training when using gradient_checkpointing. Refer to issue #9496
 2024-09-23 12:30:24 -10:00
M Saqlain
14f6464bef [Tests] Reduce the model size in the lumina test (#8985) 
* Reduced model size for lumina-tests

* Handled failing tests
 2024-09-23 20:35:50 +05:30
Sayak Paul
ba5af5aebb [Cog] some minor fixes and nits (#9466) 
* fix positional arguments in check_inputs().

* add video and latetns to check_inputs().

* prep latents_in_channels.

* quality

* multiple fixes.

* fix
 2024-09-23 11:27:05 +05:30
Sayak Paul
aa73072f1f [CI] fix nightly model tests (#9483) 
* check if default attn procs fix it.

* print

* print

* replace

* style./

* replace revision with variant.

* replace with stable-diffusion-v1-5/stable-diffusion-inpainting.

* replace with stable-diffusion-v1-5/stable-diffusion-v1-5.

* fix
 2024-09-21 07:44:47 +05:30
Aryan
e5d0a328d6 [refactor] LoRA tests (#9481) 
* refactor scheduler class usage

* reorder to make tests more readable

* remove pipeline specific checks and skip tests directly

* rewrite denoiser conditions cleaner

* bump tolerance for cog test
 2024-09-21 07:10:36 +05:30
Vladimir Mandic
14a1b86fc7 Several fixes to Flux ControlNet pipelines (#9472) 
* fix flux controlnet pipelines

---------

Co-authored-by: yiyixuxu <yixu310@gmail.com>
 2024-09-19 15:49:36 -10:00
Aryan
2b443a5d62 [training] CogVideoX Lora (#9302) 
* cogvideox lora training draft

* update

* update

* update

* update

* update

* make fix-copies

* update

* update

* apply suggestions from review

* apply suggestions from reveiw

* fix typo

* Update examples/cogvideo/train_cogvideox_lora.py

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

* fix lora alpha

* use correct lora scaling for final test pipeline

* Update examples/cogvideo/train_cogvideox_lora.py

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

* apply suggestions from review; prodigy optimizer

YiYi Xu <yixu310@gmail.com>

* add tests

* make style

* add README

* update

* update

* make style

* fix

* update

* add test skeleton

* revert lora utils changes

* add cleaner modifications to lora testing utils

* update lora tests

* deepspeed stuff

* add requirements.txt

* deepspeed refactor

* add lora stuff to img2vid pipeline to fix tests

* fight tests

* add co-authors

Co-Authored-By: Fu-Yun Wang <1697256461@qq.com>

Co-Authored-By: zR <2448370773@qq.com>

* fight lora runner tests

* import Dummy optim and scheduler only wheh required

* update docs

* add coauthors

Co-Authored-By: Fu-Yun Wang <1697256461@qq.com>

* remove option to train text encoder

Co-Authored-By: bghira <bghira@users.github.com>

* update tests

* fight more tests

* update

* fix vid2vid

* fix typo

* remove lora tests; todo in follow-up PR

* undo img2vid changes

* remove text encoder related changes in lora loader mixin

* Revert "remove text encoder related changes in lora loader mixin"

This reverts commit f8a8444487.

* update

* round 1 of fighting tests

* round 2 of fighting tests

* fix copied from comment

* fix typo in lora test

* update styling

Co-Authored-By: YiYi Xu <yixu310@gmail.com>

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: zR <2448370773@qq.com>
Co-authored-by: Fu-Yun Wang <1697256461@qq.com>
Co-authored-by: bghira <bghira@users.github.com>
 2024-09-19 14:37:57 +05:30
Sayak Paul
d13b0d63c0 [Flux] add lora integration tests. (#9353) 
* add lora integration tests.

* internal note

* add a skip marker.
 2024-09-19 09:21:28 +05:30