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
3,687 Commits 740 Branches 94 Tags
215e6804d3bb011faead6ddc8db00cdbe31b6e91
Commit Graph

3687 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Dhruv Nair
215e6804d3 Unpin torch versions in CI (#6945) 
* update

* update

* update

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-12 16:01:05 +05:30
Disty0
9254d1f39a Pass device to enable_model_cpu_offload in maybe_free_model_hooks (#6937) 2024-02-12 13:42:32 +05:30
Piyush Thakur
e1bdcc7af3 [Model Card] standardize T2I Sdxl Lora model card (#6944) 
* standardize model card template t2i-lora-sdxl

* type annotations
 2024-02-12 11:45:40 +05:30
Dhruv Nair
84905ca728 Update PixArt Alpha test module to match src module (#6943) 
update
 2024-02-12 11:01:33 +05:30
Piyush Thakur
6f336650c3 [Model Card] standardize T2I Sdxl model card (#6942) 
standardize model card template t2i-sdxl
 2024-02-12 10:01:20 +05:30
Piyush Thakur
06a042cd0e [Model Card] standardize T2I Lora model card (#6940) 
standardize model card t2i-lora
 2024-02-12 10:01:13 +05:30
Piyush Thakur
8772496586 [Model Card] standardize T2I model card (#6939) 
* standardize model card

* fix base_model
 2024-02-12 10:00:41 +05:30
dg845
35fd84be27 Replace hardcoded values in SchedulerCommonTest with properties (#5479) 
---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
 2024-02-10 21:34:20 -10:00
YiYi Xu
f2756253e6 Fix a bug in AutoPipeline.from_pipe when switching pipeline with optional components (#6820) 
* fix

* add tests

---------

Co-authored-by: yiyixuxu <yixu310@gmail,com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-09 21:56:52 -10:00
YiYi Xu
0071478d9e allow attention processors to have different signatures (#6915) 
add

Co-authored-by: yiyixuxu <yixu310@gmail,com>
 2024-02-09 21:56:19 -10:00
Sayak Paul
7c8cab313e post release 0.26.2 (#6885) 
* post release

* style

* Empty-Commit

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2024-02-09 07:36:38 -10:00
Sayak Paul
ca9ed5e8d1 [LoRA] deprecate certain lora methods from the old backend. (#6889) 
* deprecate certain lora methods from the old backend.

* uncomment necessary things.

* safe remove old lora backend 👋
 2024-02-09 17:14:32 +01:00
Bingxin Ke
98b6bee1a1 [Community Pipeline][Bug Fix] marigold_depth_estimation: input image value range (#6787) 
[FIX] IMPORTANT: rgb normalization
 2024-02-09 16:55:37 +01:00
Dhruv Nair
ab7113487c More IPAdapter test fixes (#6888) 
update
 2024-02-09 13:54:58 +05:30
camaro
59c307f1d5 Standardize model card for Controlnet (#6910) 
* controlnet

* style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-09 12:29:08 +05:30
Sayak Paul
159885adc6 correct hub_token exposition behaviour (thanks to @bghira). (#6918) 2024-02-08 18:38:27 -10:00
Aryan
7337eea59b [refactor] unnecessary lines in decode_latents in video pipelines (#6682) 
* refactor decode latents in video pipelines

* make fix-copies
 2024-02-08 17:10:52 -10:00
camaro
f07899a57c Standardize model card for Controlnet SDXL (#6908) 
controlnet-sdxl
 2024-02-09 07:53:39 +05:30
camaro
a83cc0c0bc Standardize model card for Controlnet flax (#6909) 
* controlnet-flax

* style

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-09 07:52:56 +05:30
C Q
db5194a45d Fix Compatibility Issues in stable_diffusion_xl_reference.py (#6251) 
* Fix Compatibility Issues in stable_diffusion_xl_reference.py

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-02-08 10:52:59 -10:00
shaoxiaowang
e6c9c2513f fix examples/community/pipeline_stable_diffusion_xl_instantid.py (#6759) 
Co-authored-by: wangshaoxiao <wangshaoxiao@xiaomi.com>
 2024-02-08 10:09:40 -10:00
Kyunghwan Kim
d643b6691f Add vae tiling and slicing in img2img and inpaint (#6871) 
* Add vae tiling in img2img and inpaint

* Add vae tiling not slicing
 2024-02-08 09:50:00 -10:00
Michael
f5c9be3a0a Remove <cat-toy> validation prompt from examples/textual_inversion/textual_inversion_sdxl.py (#6877) 
Remove <cat-toy> validation prompt from textual_inversion_sdxl.py

The `<cat-toy>` validation prompt is a default choice for the example task in the README. But no other part of `textual_inversion_sdxl.py` references the cat toy and `textual_inversion.py` has a default validation prompt of `None` as well.

So bring `textual_inversion_sdxl.py` in line with `textual_inversion.py` and change default validation prompt to `None`
 2024-02-08 09:46:06 -10:00
Laisky.Cai
1824d0050e fix: load_image should support PIL Image (#6904) 
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-08 08:25:37 -10:00
Sayak Paul
30e5e81d58 change to 2024 in the license (#6902) 
change to 2024
 2024-02-08 08:19:31 -10:00
Masamune Ishihara
8de78001df Add fps argument to export_to_gif function. (#6786) 2024-02-08 21:59:51 +05:30
Patryk Bartkowiak
3ac2357794 changed positional parameters to named parameters like in docs (#6905) 
Co-authored-by: Patryk Bartkowiak <patryk.bartkowiak@tcl.com>
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>
 2024-02-08 21:39:03 +05:30
Ehsan Akhgari
17808a091e Fix bug when converting checkpoint to diffusers format (#6900) 
This fixes #6899.
 2024-02-08 18:52:11 +05:30
Sayak Paul
491a933a1b [I2VGenXL] attention_head_dim in the UNet (#6872) 
* attention_head_dim

* debug

* print more info

* correct num_attention_heads behaviour

* down_block_num_attention_heads -> num_attention_heads.

* correct the image link in doc.

* add: deprecation for num_attention_head

* fix: test argument to use attention_head_dim

* more fixes.

* quality

* address comments.

* remove depcrecation.
 2024-02-08 12:30:14 +05:30
Sayak Paul
aa82df52e7 [IP Adapters] introduce ip_adapter_image_embeds in the SD pipeline call (#6868) 
* add: support for passing ip adapter image embeddings

* debugging

* make feature_extractor unloading conditioned on safety_checker

* better condition

* type annotation

* index to look into value slices

* more debugging

* debugging

* serialize embeddings dict

* better conditioning

* remove unnecessary prints.

* Update src/diffusers/loaders/ip_adapter.py

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

* make fix-copies and styling.

* styling and further copy fixing.

* fix: check_inputs call in controlnet sdxl img2img pipeline

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-02-08 11:10:10 +05:30
Srimanth Agastyaraju
a11b0f83b7 Fix: training resume from fp16 for SDXL Consistency Distillation (#6840) 
* Fix: training resume from fp16 for lcm distill lora sdxl

* Fix coding quality - run linter

* Fix 1 - shift mixed precision cast before optimizer

* Fix 2 - State dict errors by removing load_lora_into_unet

* Update train_lcm_distill_lora_sdxl.py - Revert default cache dir to None

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-08 11:09:29 +05:30
Sayak Paul
1835510524 Remove torch_dtype in to() to end deprecation (#6886) 
* remove torch_dtype from to()

* remove torch_dtype from usage scripts.

* remove old lora backend

* Revert "remove old lora backend"

This reverts commit adcddf6ba4.
 2024-02-08 09:38:57 +05:30
camaro
4a3d52850b fix: keyword argument mismatch (#6895) 2024-02-08 09:37:56 +05:30
YiYi Xu
97d004b9b4 [ip-adapter] make sure length of scale is same as number of ip-adapters when using set_ip_adapter_scale (#6884) 
add

Co-authored-by: yiyixuxu <yixu310@gmail,com>
 2024-02-07 10:13:12 -10:00
Sayak Paul
76696dca55 [Model Card] standardize dreambooth model card (#6729) 
* feat: standarize model card creation for dreambooth training.

* correct 'inference

* remove comments.

* take component out of kwargs

* style

* add: card template to have a leaner description.

* widget support.

* propagate changes to train_dreambooth_lora

* propagate changes to custom diffusion

* make widget properly type-annotated
 2024-02-07 15:07:11 +05:30
Félix Sanz
17612de451 fix: typo in callback function name and property (#6834) 
* fix: callback function name is incorrect

On this tutorial there is a function defined and then used inside `callback_on_step_end` argument, but the name was not correct (mismatch)

* fix: typo in num_timestep (correct is num_timesteps)

fixed property name
 2024-02-06 12:05:40 -08:00
Dhruv Nair
994360f7a5 Fix last IP Adapter test (#6875) 
update
 2024-02-06 08:53:40 -10:00
Dhruv Nair
e6a48db633 Refactor Deepfloyd IF tests. (#6855) 
* update

* update

* update
 2024-02-06 16:43:17 +05:30
sayakpaul
4f1df69d1a Revert "add attention_head_dim" 
This reverts commit 15f6b22466.
 2024-02-06 14:48:49 +05:30
sayakpaul
15f6b22466 add attention_head_dim 2024-02-06 14:48:07 +05:30
Sayak Paul
e6fd9ada3a [I2vGenXL] clean up things (#6845) 
* remove _to_tensor

* remove _to_tensor definition

* remove _collapse_frames_into_batch

* remove lora for not bloating the code.

* remove sample_size.

* simplify code a bit more

* ensure timesteps are always in tensor.
 2024-02-06 09:22:07 +05:30
Edward Li
493228a708 Fix AutoencoderTiny with use_slicing (#6850) 
* Fix `AutoencoderTiny` with `use_slicing`

When using slicing with AutoencoderTiny, the encoder mistakenly encodes the entire batch for every image in the batch.

* Fixed formatting issue
 2024-02-05 09:18:22 -10:00
Dhruv Nair
8bf046b7fb Add single file and IP Adapter support to PIA Pipeline (#6851) 
update
 2024-02-05 16:23:18 +05:30
Dhruv Nair
bb99623d09 Update IP Adapter tests to use cosine similarity distance (#6806) 
* update

* update
 2024-02-05 16:22:59 +05:30
Dhruv Nair
fdf55b1f1c Fix posix path issue in testing utils (#6849) 
update
 2024-02-05 08:57:18 +05:30
小咩Goat
c6f8c310c3 Fix forward pass in UNetMotionModel when gradient checkpoint is enabled (#6744) 
fix #6742

Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
 2024-02-05 08:04:01 +05:30
YiYi Xu
64909f17b7 update IP-adapter code in UNetMotionModel (#6828) 
fix

Co-authored-by: yiyixuxu <yixu310@gmail,com>
 2024-02-05 07:26:46 +05:30
Dhruv Nair
f09ca909c8 Multiple small fixes to Video Pipeline docs (#6805) 
* update

* update

* update

* Update src/diffusers/pipelines/i2vgen_xl/pipeline_i2vgen_xl.py

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

* update

* update

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
 2024-02-05 07:24:38 +05:30
YiYi Xu
a5fc62f819 add self.use_ada_layer_norm_* params back to BasicTransformerBlock (#6841) 
fix sd reference community ppeline

Co-authored-by: yiyixuxu <yixu310@gmail,com>
 2024-02-04 11:16:44 -10:00
Linoy Tsaban
fbdf26bac5 [dreambooth lora sdxl] add sdxl micro conditioning (#6795) 
* add micro conditioning

* remove redundant lines

* style

* fix missing 's'

* fix missing shape bug due to missing RGB if statement

* remove redundant if, change arg order

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2024-02-04 16:00:09 +02:00