AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-29 07:22:12 +03:00
Code Activity
3,249 Commits 742 Branches 94 Tags
beb8aa2c052d156c58bf3e73f8500c77566f2c16
Commit Graph

667 Commits

Author SHA1 Message Date
sayakpaul
cece781921 reduce lora rank. 2023-11-14 13:55:24 +05:30
sayakpaul
8c4eaf6715 autocast 2023-11-14 13:49:22 +05:30
sayakpaul
6a690abd24 autocast device determination from accelerator 2023-11-14 13:44:10 +05:30
sayakpaul
842df25c10 add test 2023-11-14 13:23:18 +05:30
sayakpaul
808f61ead9 fix 2023-11-14 11:43:00 +05:30
sayakpaul
ca076c78f0 fix serialization 2023-11-14 11:36:34 +05:30
sayakpaul
cff23edb23 remove kohya utility and make style 2023-11-14 11:33:36 +05:30
sayakpaul
53c13f7f8d remove prints. 2023-11-14 11:23:33 +05:30
sayakpaul
ba0d0f25d4 debug 2023-11-14 11:19:46 +05:30
sayakpaul
fc5546fecc debug 2023-11-14 11:12:15 +05:30
sayakpaul
3b066d2657 fix state dict serialization 2023-11-14 11:04:02 +05:30
sayakpaul
1fd3378282 feat: no separate teacher unet. 2023-11-14 10:37:38 +05:30
sayakpaul
6aa2dd8cab state dict rejigging 2023-11-14 10:29:09 +05:30
Sayak Paul
858009b401 Merge branch 'main' into lcm-lora-sdxl-datasets 2023-11-14 10:18:23 +05:30
Long(Tony) Lian
5b231aa38b Fix the pipeline name in the examples for LMD+ pipeline. Add a colab link to pipeline README. (#5775) 
* Fix the pipeline name in the examples for LMD+ pipeline

* Add LMD+ colab link

* Apply code formatting

---------

Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
 2023-11-14 07:43:37 +05:30
Thuan H. Nguyen
8fcd52febb Correct code for distributed training of RealFill (#5740) 
Correct code for distributed training
 2023-11-13 19:01:15 +01:00
Nicolas Hug
0488810f61 Fix realfill example compatibility with latest torchvision version (#5736) 2023-11-13 18:55:17 +01:00
Patrick von Platen
ef7787ea59 make style 2023-11-13 18:54:53 +01:00
Jianqi Pan
1ce4b5f3e3 fix: fix forward function signature of controlnet reference_only pipeline example (#5717) 
fix: ignore other args

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2023-11-13 18:54:21 +01:00
Younes Belkada
8789d0b6c7 fix styling issues on main (#5754) 
fix styling issues
 2023-11-13 20:05:15 +05:30
sayakpaul
e4edb31bfc style 2023-11-13 18:51:49 +05:30
sayakpaul
d7f632e658 final serialization in kohya 2023-11-13 18:42:17 +05:30
sayakpaul
6b2e42f2b9 make vae encoding batch size an arg 2023-11-13 18:14:12 +05:30
Sayak Paul
7967247113 Apply suggestions from code review 
Co-authored-by: Suraj Patil <surajp815@gmail.com>
 2023-11-13 18:10:11 +05:30
sayakpaul
3d4b1da091 no need to use use_auth_token 2023-11-13 12:55:06 +05:30
sayakpaul
dd93227ec3 fix batch size. 2023-11-13 11:36:55 +05:30
sayakpaul
df707545b7 unconditional predictions debug 2023-11-13 11:18:30 +05:30
sayakpaul
c7f28284f9 fix microconditions. 2023-11-13 10:38:28 +05:30
sayakpaul
6d2f740706 fix; dataloading. 2023-11-13 10:26:40 +05:30
sayakpaul
8c4d4b6a99 fix num_update_steps_per_epoch 2023-11-13 10:24:06 +05:30
sayakpaul
bc8cfddf78 fix batch unmarshalling 2023-11-13 10:20:04 +05:30
sayakpaul
728aa8a614 fix num_update_steps 2023-11-13 10:11:57 +05:30
sayakpaul
9e49fd2eaf remove comments. 2023-11-13 09:59:28 +05:30
sayakpaul
88efd154bd suit up the args. 2023-11-13 09:24:55 +05:30
sayakpaul
ca7f220b8a add: script to train lcm lora for sdxl with 🤗 datasets 2023-11-13 09:12:15 +05:30
Long(Tony) Lian
b1fbef544c Add LLM-grounded Diffusion (LMD+) pipeline (#5634) 
* Add LLM-grounded Diffusion (LMD+) pipeline

* Update the formatting

* Applied formatting
 2023-11-11 12:51:05 +05:30
Sayak Paul
1477865e48 post release v0.23.0 (#5730) 
* post release

* fix: variant test

* up

* fix: test
 2023-11-10 16:35:44 +05:30
aihao
1f87f83e68 add load_datasete data_dir parameter (#5747) 2023-11-09 23:03:14 -10:00
Suraj Patil
db2d8e76f8 Add LCM Scripts (#5727) 
* add lcm scripts

* Co-authored-by: dgu8957@gmail.com
 2023-11-09 17:29:12 +01:00
Sayak Paul
64603389da post release (v0.22.0) (#5658) 
post release
 2023-11-06 16:23:38 +01:00
Patrick von Platen
4f2bf67355 Revert "Fix the order of width and height of original size in SDXL training script" (#5614) 
Revert "Fix the order of width and height of original size in SDXL training script (#5382)"

This reverts commit 45db049973.
 2023-11-01 22:04:47 +01:00
M. Tolga Cangöz
442017ccc8 [Docs] Fix typos (#5583) 
* Add Copyright info

* Fix typos, improve, update

* Update deepfloyd_if.md

* Update ldm3d_diffusion.md

* Update opt_overview.md
 2023-10-31 10:04:08 -07:00
YiYi Xu
ce9484b139 fix a mistake in text2image training script for kandinsky2.2 (#5244) 
fix

Co-authored-by: yiyixuxu <yixu@Yis-MacBook-Pro.local>
 2023-10-30 23:06:16 -10:00
Jincheng Miao
ed00ead345 [Community Pipelines] add textual inversion support for stable_diffusion_ipex (#5571) 2023-10-31 11:54:16 +05:30
Thuan H. Nguyen
5b087e82d1 Add realfill (#5456) 
* Add realfill

* Move realfill folder

* Fix some format issues
 2023-10-30 15:21:40 +01:00
jiaqiw09
e140c0562e fix error reported 'find_unused_parameters' running in mutiple GPUs (#5355) 
* fix error reported 'find_unused_parameters' running in mutiple GPUs or NPUs

* fix code check of importing module by its alphabetic order

---------

Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
 2023-10-27 22:49:14 +05:30
nickkolok
0d4b459be6 Update train_dreambooth.py - fix typos (#5539) 2023-10-26 13:35:05 -07:00
Ran Ran
8959c5b9de Add from_pt flag to enable model from PT (#5501) 
* Add from_pt flag to enable model from PT

* Format the file

* Reformat the file
 2023-10-25 23:07:34 +02:00
Patrick von Platen
d420d71398 make style 2023-10-25 16:12:14 +02:00
Logan
a1fad8286f Add a new community pipeline (#5477) 
* Add a new community pipeline

examples/community/latent_consistency_img2img.py

which can be called like this

import torch
from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(
                "SimianLuo/LCM_Dreamshaper_v7", custom_pipeline="latent_consistency_txt2img", custom_revision="main")

            # To save GPU memory, torch.float16 can be used, but it may compromise image quality.
pipe.to(torch_device="cuda", torch_dtype=torch.float32)

img2img=LatentConsistencyModelPipeline_img2img(
    vae=pipe.vae,
    text_encoder=pipe.text_encoder,
    tokenizer=pipe.tokenizer,
    unet=pipe.unet,
    #scheduler=pipe.scheduler,
    scheduler=None,
    safety_checker=None,
    feature_extractor=pipe.feature_extractor,
    requires_safety_checker=False,
)

img = Image.open("thisismyimage.png")

result = img2img(prompt,img,strength,num_inference_steps=4)

* Apply suggestions from code review

Fix name formatting for scheduler

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

* update readme (and run formatter on latent_consistency_img2img.py)

---------

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2023-10-25 16:11:56 +02:00