1
0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00

2647 Commits

Author SHA1 Message Date
Sayak Paul
965e52ce61 Release: v0.19.2 v0.19.2 2023-07-28 23:33:51 +05:30
Sayak Paul
b1e52794a2 [Feat] Support SDXL Kohya-style LoRA (#4287)
* sdxl lora changes.

* better name replacement.

* better replacement.

* debugging

* debugging

* debugging

* debugging

* debugging

* remove print.

* print state dict keys.

* print

* distingisuih better

* debuggable.

* fxi: tyests

* fix: arg from training script.

* access from class.

* run style

* debug

* save intermediate

* some simplifications for SDXL LoRA

* styling

* unet config is not needed in diffusers format.

* fix: dynamic SGM block mapping for SDXL kohya loras (#4322)

* Use lora compatible layers for linear proj_in/proj_out (#4323)

* improve condition for using the sgm_diffusers mapping

* informative comment.

* load compatible keys and embedding layer maaping.

* Get SDXL 1.0 example lora to load

* simplify

* specif ranks and hidden sizes.

* better handling of k rank and hidden

* debug

* debug

* debug

* debug

* debug

* fix: alpha keys

* add check for handling LoRAAttnAddedKVProcessor

* sanity comment

* modifications for text encoder SDXL

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* denugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* up

* up

* up

* up

* up

* up

* unneeded comments.

* unneeded comments.

* kwargs for the other attention processors.

* kwargs for the other attention processors.

* debugging

* debugging

* debugging

* debugging

* improve

* debugging

* debugging

* more print

* Fix alphas

* debugging

* debugging

* debugging

* debugging

* debugging

* debugging

* clean up

* clean up.

* debugging

* fix: text

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Batuhan Taskaya <batuhan@python.org>
2023-07-28 23:28:22 +05:30
Patrick von Platen
c3e3a1ee10 [SDXL] Make watermarker optional under certain circumstances to improve usability of SDXL 1.0 (#4346)
* improve sdxl

* more fixes

* improve sdxl

* improve sdxl

* improve sdxl

* finish
2023-07-28 23:28:11 +05:30
Patrick von Platen
9cde56a729 [ONNX] Don't download ONNX model by default (#4338)
* [Download] Don't download ONNX weights by default

* [Download] Don't download ONNX weights by default

* [Download] Don't download ONNX weights by default

* fix more

* finish

* finish

* finish
2023-07-28 23:27:58 +05:30
Patrick von Platen
c63d7cdba0 [SDXL Refiner] Fix refiner forward pass for batched input (#4327)
* fix_batch_xl

* Fix other pipelines as well

* up

* up

* Update tests/pipelines/stable_diffusion_xl/test_stable_diffusion_xl_inpaint.py

* sort

* up

* Finish it all up Co-authored-by: Bagheera <bghira@users.github.com>

* Co-authored-by: Bagheera bghira@users.github.com

* Co-authored-by: Bagheera <bghira@users.github.com>

* Finish it all up Co-authored-by: Bagheera <bghira@users.github.com>
2023-07-28 23:27:36 +05:30
Patrick von Platen
aa4634a7fa Release: v0.19.1 v0.19.1 2023-07-27 20:00:43 +02:00
Patrick von Platen
0709650e9d [Local loading] Correct bug with local files only (#4318)
* [Local loading] Correct bug with local files only

* file not found error

* fix

* finish
2023-07-27 20:00:21 +02:00
YiYi Xu
a9829164f4 fix a bug in StableDiffusionUpscalePipeline when prompt is None (#4278)
* fix batch_size

* add test

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
2023-07-27 20:00:12 +02:00
Duong A. Nguyen
49c95178ad Fix SDXL conversion from original to diffusers (#4280)
* fix sdxl conversion

* convention
2023-07-27 20:00:02 +02:00
Patrick von Platen
c2f755bc62 [Torch.compile] Fixes torch compile graph break (#4315)
* fix torch compile

* Fix all

* make style
2023-07-27 19:59:55 +02:00
YiYi Xu
2fb877b66c update Kandinsky doc (#4301)
* update doc

* fix an error in autopipe doc

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
2023-07-27 19:59:50 +02:00
Patrick von Platen
ef9824f9f7 Release: v0.19.0 v0.19.0 2023-07-26 21:03:45 +02:00
Patrick von Platen
5623ea065a quick fix 2023-07-26 21:01:17 +02:00
Patrick von Platen
16049caf79 quick fix 2023-07-26 18:47:21 +00:00
Patrick von Platen
6a6dfe1cbd Rename (#4294)
* up

* Apply suggestions from code review

* Apply suggestions from code review

* up
2023-07-26 20:41:21 +02:00
Ella Charlaix
b83bdce42a add openvino and onnx runtime SD XL documentation (#4285)
* add openvino SD XL documentation

* add onnx SD XL integration

* rephrase

* update doc

* add images

* update model
2023-07-26 20:25:07 +02:00
camenduru
c6ae9b7df6 Where did this 'x' come from, Elon? (#4277)
* why mdx?

* why mdx?

* why mdx?

* no x for kandinksy either

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2023-07-26 18:18:14 +02:00
Patrick von Platen
b3e5cd6b4d [Kandinsky] Add combined pipelines / Fix cpu model offload / Fix inpainting (#4207)
* Add combined pipeline

* Download readme

* Upload

* up

* up

* fix final

* Add enable model cpu offload kandinsky

* finish

* finish

* Fix

* fix more

* make style

* fix kandinsky mask

* fix inpainting test

* add callbacks

* add tests

* fix tests

* Apply suggestions from code review

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

* docs

* docs

* correct docs

* fix tests

* add warning

* correct docs

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
2023-07-26 17:13:55 +02:00
Patrick von Platen
b37dc3b3cd Fix all missing optional import statements from pipeline folders (#4272)
* fix circular import

* fix imports when watermark not specified

* fix all pipelines
2023-07-26 01:46:05 +02:00
Batuhan Taskaya
ff8f58086b Load Kohya-ss style LoRAs with auxilary states (#4147)
* Support to load Kohya-ss style LoRA file format (without restrictions)

Co-Authored-By: Takuma Mori <takuma104@gmail.com>
Co-Authored-By: Sayak Paul <spsayakpaul@gmail.com>

* tmp: add sdxl to mlp_modules

---------

Co-authored-by: Takuma Mori <takuma104@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2023-07-26 00:24:19 +02:00
Sayak Paul
161449d51a [SDXL DreamBooth LoRA] multiple fixes (#4262)
* add automatic licensing.

* debugging

* debugging

* more debugging

* more debugging.

* run make fix-copies.

* change to default tracker.
2023-07-25 21:10:01 +02:00
Steven Liu
34abee0907 [docs] Fix image in SDXL docs (#4267)
fix image link
2023-07-25 09:41:11 -07:00
Harutatsu Akiyama
428dbfecd9 [SDXL and IP2P]: instruction pix2pix XL training and pipeline (#4079)
* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* [Community] Implementation of the IADB community pipeline (#3996)

* community pipeline: implementation of iadb

* iadb.py: reformat using black

* iadb.py: linting update

* add kandinsky to readme table (#4081)

Co-authored-by: yiyixuxu <yixu310@gmail,com>

* [From Single File] Force accelerate to be installed (#4078)

force accelerate to be installed

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Support instruction pix2pix sdxl

* Clean up IP2P SDXL code

* Clean up IP2P SDXL code

* [IP2P and SDXL] clean up code

* [IP2P and SDXL] clean up code

* [IP2P and SDXL] clean up code

* [IP2P SDXL] Address code reviews

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews, add docs, tests

* [IP2P SDXL] Address code reviews

* [IP2P SDXL] Address code reviews

* [IP2P SDXL] Add README_SDXL

* [IP2P SDXL] Address code reviews

* [IP2P SDXL] Address code reviews

* [IP2P SDXL] Fix the copy problems

* [IP2P SDXL] Add license

* [IP2P SDXL] Add license

* [IP2P SDXL] Add license

* [IP2P SDXL] Address code reivew for selecting VAE andd others

* [IP2P SDXL] Update README_sdxl

* [IP2P SDXL] Update __init__

* [IP2P SDXL] Update dummy_torch_and_transformers_and_invisible_watermark_objects

* address patrick's comments and some additions to readmes.

---------

Co-authored-by: Harutatsu Akiyama <kf.zy.qin@gmail.com>
Co-authored-by: Thomas Chambon <36728882+tchambon@users.noreply.github.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2023-07-25 18:19:35 +05:30
Ragnar Rova
4e2a021829 Model path for sdxl wrong in dreambooth README (#4261) 2023-07-25 18:06:50 +05:30
Patrick von Platen
ebfe343149 [from_single_file] Fix circular import (#4259)
* up

* finish

* fix final
2023-07-25 14:30:39 +02:00
Sayak Paul
5ef6b8fa53 Update README_sdxl.md to change the note on default hyperparameters (#4258) 2023-07-25 16:57:48 +05:30
YiYi Xu
c11d11d63d [draft v2] AutoPipeline (#4138)
* initial

* style

* from ...pipelines -> from ..pipeline_util

* make style

* fix-copies

* fix value_guided_sampling oops

* style

* add test

* Show failing test

* update from_pipe

* fix

* add controlnet, additional test and register unused original config

* update for controlnet

* Apply suggestions from code review

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* store unused config as private attribute and pass if can

* add doc

* kandinsky inpaint pipeline does not work with decoder checkpoint

* update doc

* Apply suggestions from code review

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

* style

* Apply suggestions from code review

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* fix

* Apply suggestions from code review

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2023-07-25 13:20:35 +02:00
Patrick von Platen
d74561da2c [SDXL] Improve docs (#4196)
* Improve docs

* Correct docs

* Add better example inpaint

* make style

* Apply suggestions from code review

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

* fix

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
2023-07-25 12:48:25 +02:00
Patrick von Platen
a0422ed0c9 [From Single File] Allow vae to be loaded (#4242)
* Allow vae to be loaded

* up
2023-07-25 12:16:43 +02:00
Will Berman
3dd339379d do not pass list to accelerator.init_trackers (#4248) 2023-07-24 21:10:37 -07:00
nupurkmr9
5652c43f83 Resolve bf16 error as mentioned in this [issue](https://github.com/huggingface/diffusers/issues/4139#issuecomment-1639977304) (#4214)
* resolve bf16 error

* resolve bf16 error

* resolve bf16 error

* resolve bf16 error

* resolve bf16 error

* resolve bf16 error

* resolve bf16 error
2023-07-25 05:41:19 +05:30
Sayak Paul
365e8461ac [SDXL DreamBooth LoRA] add support for text encoder fine-tuning (#4097)
* Allow low precision sd xl

* finish

* finish

* feat: initial draft for supporting text encoder lora finetuning for SDXL DreamBooth

* fix: variable assignments.

* add: autocast block.

* add debugging

* vae dtype hell

* fix: vae dtype hell.

* fix: vae dtype hell 3.

* clean up

* lora text encoder loader.

* fix: unwrapping models.

* add: tests.

* docs.

* handle unexpected keys.

* fix vae dtype in the final inference.

* fix scope problem.

* fix: save_model_card args.

* initialize: prefix to None.

* fix: dtype issues.

* apply gixes.

* debgging.

* debugging

* debugging

* debugging

* debugging

* debugging

* add: fast tests.

* pre-tokenize.

* address: will's comments.

* fix: loader and tests.

* fix: dataloader.

* simplify dataloader.

* length.

* simplification.

* make style && make quality

* simplify state_dict munging

* fix: tests.

* fix: state_dict packing.

* Apply suggestions from code review

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2023-07-25 05:35:48 +05:30
Sayak Paul
fed12376c5 [ControlNet SDXL training] fixes in the training script (#4223)
* fix: #4206

* add: sdxl controlnet training smoketest.

* remove unnecessary token inits.

* add: licensing to model card.

* include SDXL licensing in the model card and make public visibility default

* debugging

* debugging

* disable local file download.

* fix: training test.

* fix: ckpt prefix.
2023-07-25 05:31:48 +05:30
Patrick von Platen
95b7de88fd [Docs] Fix from pretrained docs (#4240)
* [Docs] Fix from pretrained docs

* [Docs] Fix from pretrained docs
2023-07-24 20:24:29 +02:00
Apoorva Kulkarni
cbb1ead60b docs: Add missing import statement in textual_inversion inference example (#4227)
docs: Add missing import statement in textual_inversion inference instructions
2023-07-24 11:07:53 -07:00
Steven Liu
5470a4fce3 [docs] Other modalities (#4205)
remove coming soon, rl pipeline
2023-07-24 10:51:24 -07:00
39th president of the United States, probably
e98fabc550 Allow specifying denoising_start and denoising_end as integers representing the discrete timesteps, fixing the XL ensemble not working for many schedulers (#4115)
* Fix the XL ensemble not working for any kerras scheduler sigmas and having an off by one bug

* Update src/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py

* make sytle

---------

Co-authored-by: Jimmy <39@🇺🇸.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2023-07-24 19:44:35 +02:00
Cris
fa356bd4da [docs] Changed path for ControlNet in docs (#4215)
docs: changed path for control net
2023-07-24 10:13:10 -07:00
Patrick von Platen
3ba36f97b8 [SD-XL] Fix sdxl controlnet inference (#4238)
* Fix controlnet xl inference

* correct some sd xl control inference
2023-07-24 18:43:35 +02:00
Patrick von Platen
b288684d25 [SDXL] Fix sd xl encode prompt (#4237)
* [SDXL] Fix sd xl encode prompt

* add tests
2023-07-24 18:37:07 +02:00
Lucain
06eda5b232 Raise initial HTTPError if pipeline is not cached locally (#4230)
* Raise initial HTTPError if pipeline is not cached locally

* make style
2023-07-24 15:35:16 +02:00
Hu Ye
8e5921cac1 fix a bug of prompt embeds in sdxl (#4099)
* fix bug in sdxl

* Update pipeline_stable_diffusion_xl_img2img.py

* Update pipeline_stable_diffusion_xl.py

* Update pipeline_stable_diffusion_xl_img2img.py

* Update pipeline_stable_diffusion_xl_inpaint.py

* Update pipeline_stable_diffusion_xl.py

* Update pipeline_stable_diffusion_xl_img2img.py

* Update pipeline_stable_diffusion_xl_inpaint.py

* Update pipeline_stable_diffusion_xl_img2img.py

* Update pipeline_controlnet_sd_xl.py

* Update pipeline_controlnet_sd_xl.py

* Update pipeline_stable_diffusion_xl.py

* Update pipeline_stable_diffusion_xl_img2img.py

* Update pipeline_stable_diffusion_xl_inpaint.py

* Update test_stable_diffusion_xl.py

* Update test_stable_diffusion_xl.py

* Update test_stable_diffusion_xl.py

add test on prompt_embeds

* add test on prompt_embeds

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2023-07-24 10:14:20 +02:00
YiYi Xu
8e8954bd15 fix no CFG for kandinsky pipelines (#4193)
* fix bug when no cfg

* style

* fix no cfg for shap-e and cycle

* style

* fix no cfg for sdxl

* fix copies

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
2023-07-24 10:10:14 +02:00
Jackmin801
09fab5610e [fix] network_alpha when loading unet lora from old format (#4221)
fix: missed network_alpha when loading lora from old format
2023-07-24 06:13:32 +05:30
Apoorva Kulkarni
2e53936c97 docs: Typo in dreambooth example README.md (#4203)
fix: Typo in dreambooth example README.md
2023-07-21 15:16:38 -07:00
Steven Liu
a69754bb87 [docs] Clean up pipeline apis (#3905)
* start with stable diffusion

* fix

* finish stable diffusion pipelines

* fix path to pipeline output

* fix flax paths

* fix copies

* add up to score sde ve

* finish first pass of pipelines

* fix copies

* second review

* align doc titles

* more review fixes

* final review
2023-07-21 11:01:34 -07:00
Kadir Nar
bcc570b910 📄 Renamed File for Better Understanding (#4056)
* 📄 Renamed File for Better Understanding

Renamed the 'rl' file to 'run_locomotion'. This change was made to improve the clarity and readability of the codebase. The 'rl' name was ambiguous, and 'run_locomotion' provides a more clear description of the file's purpose.

Thanks 🙌

* 📁 [Docs] Renamed Directory for Better Clarity

Renamed the 'rl' directory to 'reinforcement_learning'. This change provides a clearer understanding of the directory's purpose and its contents.

* Update examples/reinforcement_learning/README.md

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* 📝 Update README

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2023-07-21 09:08:27 -07:00
Sayak Paul
4dcab9227a [SDXL ControlNet Training] Follow-up fixes (#4188)
* hash computation. thanks to @lhoestq

* disable dtype casting.

* remove comments.
2023-07-21 20:55:33 +05:30
apolinário
aed30dff6b Allow passing different prompts to each text_encoder on stable_diffusion_xl pipelines (#4156)
* sdxl prompt2

* Improve checks

* doc linting

* whoops

* remove cat

* Apply suggestions from code review

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Add other pipelines and tests

* Add multi-prompting to docs

* doc and copies check

* Fix copied froms

* Apply suggestions from code review

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

* Bring back the original code for unrelated files

* Fix tests

* Fix img2img

* Fix all

* fix

---------

Co-authored-by: multimodalart <joaopaulo.passos+multimodal@gmail.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
2023-07-21 14:50:22 +02:00
statelesshz
e2bbaa4f54 make enable_sequential_cpu_offload more generic for third-party devices (#4191)
* make enable_sequential_cpu_offload more generic for third-party devices

* make style
2023-07-21 17:15:09 +05:30