Commits · fbf6b856cc61fd22ad8635547bff4aafe05723f3 · renzhc / diffusers_dcu

"git@developer.sourcefind.cn:modelzoo/resnet50_tensorflow.git" did not exist on "38385b0a0260609192423e371ce8a30a9f5571ba"

04 Mar, 2025 1 commit

Fix incorrect seed initialization when args.seed is 0 (#10964) · b8215b1c

Alexey Zolotenkov authored Mar 04, 2025



* Fix seed initialization to handle args.seed = 0 correctly

* Apply style fixes

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

b8215b1c

24 Feb, 2025 1 commit
- [Fix] fp16 unscaling in train_dreambooth_lora_sdxl (#10889) · 170833c2
  SahilCarterr authored Feb 24, 2025
```
Fix fp16 bug
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
```
  170833c2
20 Feb, 2025 1 commit

[LoRA] add LoRA support to Lumina2 and fine-tuning script (#10818) · f10d3c6d

Sayak Paul authored Feb 20, 2025

* feat: lora support for Lumina2.

* fix-copies.

* updates

* updates

* docs.

* fix

* add: training script.

* tests

* updates

* updates

* major updates.

* updates

* fixes

* docs.

* updates

* updates

f10d3c6d

06 Feb, 2025 1 commit

[bugfix] NPU Adaption for Sana (#10724) · cd0a4a82

Leo Jiang authored Feb 06, 2025



* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* NPU Adaption for Sanna

* [bugfix]NPU Adaption for Sanna

---------
Co-authored-by: J石页 <jiangshuo9@h-partners.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

cd0a4a82

27 Jan, 2025 1 commit
- [training] Convert to ImageFolder script (#10664) · 41571773
  hlky authored Jan 27, 2025
```
* [training] Convert to ImageFolder script

* make
```
  41571773
24 Jan, 2025 1 commit

NPU Adaption for Sanna (#10409) · 07860f99

Leo Jiang authored Jan 24, 2025



* NPU Adaption for Sanna


---------
Co-authored-by: J石页 <jiangshuo9@h-partners.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

07860f99

21 Jan, 2025 3 commits

Remove the FP32 Wrapper when evaluating (#10617) · 158a5a87

Muyang Li authored Jan 21, 2025



Remove the FP32 Wrapper
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

158a5a87

Enable dreambooth lora finetune example on other devices (#10602) · 012d08b1

jiqing-feng authored Jan 21, 2025



* enable dreambooth_lora on other devices
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

* enable xpu
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

* check cuda device before empty cache
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

* fix comment
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

* import free_memory
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

---------
Signed-off-by: jiqing-feng <jiqing.feng@intel.com>

012d08b1

[chore] change licensing to 2025 from 2024. (#10615) · 4ace7d04
Sayak Paul authored Jan 21, 2025
```
change licensing to 2025 from 2024.
```
4ace7d04

15 Jan, 2025 1 commit
- [Sana 4K] Add vae tiling option to avoid OOM (#10583) · b0c89738
  Leo Jiang authored Jan 15, 2025
```
Co-authored-by: J石页 <jiangshuo9@h-partners.com>
```
  b0c89738
30 Dec, 2024 1 commit

[training] add ds support to lora sd3. (#10378) · 5f724735

Sayak Paul authored Dec 30, 2024



* add ds support to lora sd3.
Co-authored-by: leisuzz <jiangshuonb@gmail.com>

* style.

---------
Co-authored-by: leisuzz <jiangshuonb@gmail.com>
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

5f724735

23 Dec, 2024 2 commits

[chore] post release 0.32.0 (#10361) · 92933ec3
Sayak Paul authored Dec 24, 2024
```
* post release 0.32.0

* stylew
```
92933ec3

[SANA LoRA] sana lora training tests and misc. (#10296) · 76e2727b

Sayak Paul authored Dec 23, 2024



* sana lora training tests and misc.

* remove push to hub

* Update examples/dreambooth/train_dreambooth_lora_sana.py
Co-authored-by: Aryan <aryan@huggingface.co>

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

76e2727b

19 Dec, 2024 1 commit
- [chore] Update README_sana.md to update the default model (#10285) · 9c0e20de
  Sayak Paul authored Dec 19, 2024
```
Update README_sana.md to update the default model
```
  9c0e20de
18 Dec, 2024 2 commits

[chore] fix: reamde -> readme (#10276) · 63cdf9c0
Sayak Paul authored Dec 18, 2024
```
fix: reamde -> readme
```
63cdf9c0

[LoRA] feat: lora support for SANA. (#10234) · 9408aa2d

Sayak Paul authored Dec 18, 2024



* feat: lora support for SANA.

* make fix-copies

* rename test class.

* attention_kwargs -> cross_attention_kwargs.

* Revert "attention_kwargs -> cross_attention_kwargs."

This reverts commit 23433bf9bccc12e0f2f55df26bae58a894e8b43b.

* exhaust 119 max line limit

* sana lora fine-tuning script.

* readme

* add a note about the supported models.

* Apply suggestions from code review
Co-authored-by: Aryan <aryan@huggingface.co>

* style

* docs for attention_kwargs.

* remove lora_scale from pag pipeline.

* copy fix

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

9408aa2d

12 Dec, 2024 1 commit

fix min-snr implementation (#8466) · 26e80e01

Ethan Smith authored Dec 11, 2024

* fix min-snr implementation

https://github.com/kohya-ss/sd-scripts/blob/main/library/custom_train_functions.py#L66



* Update train_dreambooth.py

fix variable name mse_loss_weights

* fix divisor

* make style

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

26e80e01

25 Nov, 2024 1 commit
- Add prompt about wandb in examples/dreambooth/readme. (#10014) · 074e1235
  SkyCol authored Nov 25, 2024
```
Add files via upload
```
  074e1235
24 Nov, 2024 1 commit

[SD3 dreambooth lora] smol fix to checkpoint saving (#9993) · c4b5d2ff

Linoy Tsaban authored Nov 24, 2024

* smol change to fix checkpoint saving & resuming (as done in train_dreambooth_sd3.py)

* style

* modify comment to explain reasoning behind hidden size check

c4b5d2ff

19 Nov, 2024 1 commit

[advanced flux training] bug fix + reduce memory cost as in #9829 (#9838) · acf479bd

Linoy Tsaban authored Nov 18, 2024

* memory improvement as done here: https://github.com/huggingface/diffusers/pull/9829



* fix bug

* fix bug

* style

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

acf479bd

08 Nov, 2024 1 commit

[FIX] Fix TypeError in DreamBooth SDXL when use_dora is False (#9879) · 9cc96a64

SahilCarterr authored Nov 09, 2024



* fix use_dora

* fix style and quality

* fix use_dora with peft version

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

9cc96a64

06 Nov, 2024 1 commit
- Updated _encode_prompt_with_clip and encode_prompt in train_dreamboth_sd3 (#9800) · 76b7d86a
  SahilCarterr authored Nov 06, 2024
```
* updated encode prompt and clip encod prompt


---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
```
  76b7d86a
01 Nov, 2024 3 commits

Reduce Memory Cost in Flux Training (#9829) · a98a839d

Leo Jiang authored Nov 01, 2024



* Improve NPU performance

* Improve NPU performance

* Improve NPU performance

* Improve NPU performance

* [bugfix] bugfix for npu free memory

* [bugfix] bugfix for npu free memory

* [bugfix] bugfix for npu free memory

* Reduce memory cost for flux training process

---------
Co-authored-by: 蒋硕 <jiangshuo9@h-partners.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

a98a839d

Handling mixed precision for dreambooth flux lora training (#9565) · 3deed729

Boseong Jeon authored Nov 01, 2024



Handling mixed precision and add unwarp
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

3deed729

NPU Adaption for FLUX (#9751) · 9dcac830

Leo Jiang authored Oct 31, 2024



* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

* NPU implementation for FLUX

---------
Co-authored-by: 蒋硕 <jiangshuo9@h-partners.com>

9dcac830

31 Oct, 2024 1 commit

[training] use the lr when using 8bit adam. (#9796) · 8ce37ab0

Sayak Paul authored Oct 31, 2024



* use the lr when using 8bit adam.

* remove lr as we pack it in params_to_optimize.

---------
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

8ce37ab0

28 Oct, 2024 3 commits

[flux dreambooth lora training] make LoRA target modules configurable + small bug fix (#9646) · 743a5697

Linoy Tsaban authored Oct 28, 2024

* make lora target modules configurable and change the default

* style

* make lora target modules configurable and change the default

* fix bug when using prodigy and training te

* fix mixed precision training as  proposed in https://github.com/huggingface/diffusers/pull/9565

 for full dreambooth as well

* add test and notes

* style

* address sayaks comments

* style

* fix test

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

743a5697

[SD 3.5 Dreambooth LoRA] support configurable training block & layers (#9762) · db5b6a96

Linoy Tsaban authored Oct 28, 2024

* configurable layers

* configurable layers

* update README

* style

* add test

* style

* add layer test, update readme, add nargs

* readme

* test style

* remove print, change nargs

* test arg change

* style

* revert nargs 2/2

* address sayaks comments

* style

* address sayaks comments

db5b6a96

[Fix] remove setting lr for T5 text encoder when using prodigy in flux... · 493aa743

Biswaroop authored Oct 28, 2024


[Fix] remove setting lr for T5 text encoder when using prodigy in flux dreambooth lora script (#9473)

* fix: removed setting of text encoder lr for T5 as it's not being tuned

* fix: removed setting of text encoder lr for T5 as it's not being tuned

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

493aa743

25 Oct, 2024 1 commit
- [refactor] enhance readability of flux related pipelines (#9711) · 73b59f52
  Ina authored Oct 26, 2024
```
* flux pipline: readability enhancement.
```
  73b59f52
23 Oct, 2024 1 commit
- [SD3-5 dreambooth lora] update model cards (#9749) · bfa0aa4f
  Linoy Tsaban authored Oct 23, 2024
```
* improve readme

* style

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
```
  bfa0aa4f
22 Oct, 2024 1 commit
- post-release 0.31.0 (#9742) · e45c25d0
  Sayak Paul authored Oct 22, 2024
```
* post-release

* style
```
  e45c25d0
16 Oct, 2024 1 commit

[SD3 dreambooth-lora training] small updates + bug fixes (#9682) · ee4ab238

Linoy Tsaban authored Oct 16, 2024

* add latent caching + smol updates

* update license

* replace with free_memory

* add --upcast_before_saving to allow saving transformer weights in lower precision

* fix models to accumulate

* fix mixed precision issue as proposed in https://github.com/huggingface/diffusers/pull/9565



* smol update to readme

* style

* fix caching latents

* style

* add tests for latent caching

* style

* fix latent caching

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

ee4ab238

15 Oct, 2024 1 commit
- Dreambooth lora flux bug 3dtensor to 2dtensor (#9653) · dccf39f0
  0x名無し authored Oct 15, 2024
```
* fixed issue #9350, Tensor is deprecated

* ran make style
```
  dccf39f0
28 Sep, 2024 1 commit
- [chore] fix: retain memory utility. (#9543) · 8e7d6c03
  Sayak Paul authored Sep 28, 2024
```
* fix: retain memory utility.

* fix

* quality

* free_memory.
```
  8e7d6c03
16 Sep, 2024 1 commit

[docs] Replace runwayml/stable-diffusion-v1-5 with Lykon/dreamshaper-8 (#9428) · b52119ae

suzukimain authored Sep 17, 2024



* [docs] Replace runwayml/stable-diffusion-v1-5 with Lykon/dreamshaper-8

Updated documentation as runwayml/stable-diffusion-v1-5 has been removed from Huggingface.

* Update docs/source/en/using-diffusers/inpaint.md
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Replace with stable-diffusion-v1-5/stable-diffusion-v1-5

* Update inpaint.md

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

b52119ae

15 Sep, 2024 1 commit

[Flux Dreambooth lora] add latent caching (#9160) · 37e3603c

Linoy Tsaban authored Sep 15, 2024

* add ostris trainer to README & add cache latents of vae

* add ostris trainer to README & add cache latents of vae

* style

* readme

* add test for latent caching

* add ostris noise scheduler
https://github.com/ostris/ai-toolkit/blob/9ee1ef2a0a2a9a02b92d114a95f21312e5906e54/toolkit/samplers/custom_flowmatch_sampler.py#L95



* style

* fix import

* style

* fix tests

* style

* --change upcasting of transformer?

* update readme according to main

* keep only latent caching

* add configurable param for final saving of trained layers- --upcast_before_saving

* style

* Update examples/dreambooth/README_flux.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update examples/dreambooth/README_flux.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* use clear_objs_and_retain_memory from utilities

* style

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

37e3603c

14 Sep, 2024 1 commit

Fix the issue on sd3 dreambooth w./w.t. lora training (#9419) · e2ead7cd

Leo Jiang authored Sep 14, 2024



* Fix dtype error

* [bugfix] Fixed the issue on sd3 dreambooth training

* [bugfix] Fixed the issue on sd3 dreambooth training

---------
Co-authored-by: 蒋硕 <jiangshuo9@h-partners.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

e2ead7cd

11 Sep, 2024 1 commit
- [Tests] fix some fast gpu tests. (#9379) · adf1f911
  Sayak Paul authored Sep 11, 2024
```
fix some fast gpu tests.
```
  adf1f911
05 Sep, 2024 1 commit
- improve README for flux dreambooth lora (#9290) · 55ac421f
  Linoy Tsaban authored Sep 05, 2024
```
* improve readme

* improve readme

* improve readme

* improve readme
```
  55ac421f