GitLab

* post release v0.28.0

* style

Fixed few docstrings according to the Google Style Guide.

* fixed vae loading issue #7619

* rerun make style && make quality

* bring back model_has_vae and add change \ to / in config_file_name on windows os to make match work

* add missing import platform

* bring back import model_info

* make config_file_name OS independent

* switch to using Path.as_posix() to resolve OS dependence

* improve style

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: bssrdf <bssrdf@gmail.com>
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>

* 7529 do not disable autocast for cuda devices

* Remove typecasting error check for non-mps platforms, as a correct autocast implementation makes it a non-issue

* add autocast fix to other training examples

* disable native_amp for dreambooth (sdxl)

* disable native_amp for pix2pix (sdxl)

* remove tests from remaining files

* disable native_amp on huggingface accelerator for every training example that uses it

* convert more usages of autocast to nullcontext, make style fixes

* make style fixes

* style.

* Empty-Commit

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

* apple mps: training support for SDXL LoRA

* sdxl: support training lora, dreambooth, t2i, pix2pix, and controlnet on apple mps

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

* add kohya conversion dict.

* update readme

* typo

* add filename

* post-release

* quality

switch to logger.warning

* fix: save_model_card utility.

* fix a little more to make it more lenient.

* remove lower()

fix: prior preservation setting in DreamBooth LoRA SDXL script.
Co-authored-by: Linoy Tsaban <57615435+linoytsaban@users.noreply.github.com>

* dora in canonical script

* add mention of DoRA to readme

* add tags for diffusers training

* add tags for diffusers training

* add tags for diffusers training

* add tags for diffusers training

* add tags for diffusers training

* add tags for diffusers training

* add dora tags for drambooth lora scripts

* style

* add: dreambooth lora script for Playground v2.5

* fix: kwarg

* address suraj's comments.

* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* apply suraj's suggestion

* incorporate changes in the canonical script./

* tracker naming

* fix: schedule determination

* add: two simple tests

* remove playground script

* note about edm-style training

* address pedro's comments.

* address part of Suraj's comments.

* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* remove guidance_scale.

* use mse_loss.

* add comments for preconditioning.

* quality

* Update examples/dreambooth/train_dreambooth_lora_sdxl.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* tackle v-pred.

* Empty-Commit

* support edm for sdxl too.

* address suraj's comments.

* Empty-Commit

---------
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* fix kwarg

* Empty-Commit

* modularize log validation

* run make style

* revert import wandb

* fix code quality & import wandb

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

* modulize log validation

* run make style and refactor wanddb support

* remove redundant initialization

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

* Updated lora inference instructions

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

* Update README.md

* Update README.md

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

* post release

* style

* Empty-Commit

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

change to 2024

* 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

* 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>

Update train_dreambooth_lora_sdxl.py

* fix training resume

* update

* update

create a utility for casting the lora params during training.

* fix: training resume from fp16.

* add: comment

* remove residue from another branch.

* remove more residues.

* thanks to Younes; no hacks.

* style.

* clean things a bit and modularize _set_state_dict_into_text_encoder

* add comment about the fix detailed.

* support compile

* make style

* move unwrap_model inside function

* change unwrap call

* run make style

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

* Revert "Update examples/dreambooth/train_dreambooth.py"

This reverts commit 70ab09732e7cfec0b19c497f823ddd1c8259dad0.

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

support compile

make checkpointing compatible when using torch.compile.

* make it torch.compile comaptible

* make the text encoder compatible too.

* style

* post release

* style

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

Update README_sdxl.md

* add to dreambooth lora.

* add: t2i lora.

* add: sdxl t2i lora.

* style

* lcm lora sdxl.

* unwrap

* fix: enable_adapters().

* spit diffusers-native format from the get go.

* rejig the peft_to_diffusers mapping.

* remove validation args from textual onverson tests

* reduce number of train steps in textual inversion tests

* fix: directories.

* debig

* fix: directories.

* remove validation tests from textual onversion

* try reducing the time of test_text_to_image_checkpointing_use_ema

* fix: directories

* speed up test_text_to_image_checkpointing

* speed up test_text_to_image_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* fix

* speed up test_instruct_pix2pix_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* set checkpoints_total_limit to 2.

* test_text_to_image_lora_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints speed up

* speed up test_unconditional_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* debug

* fix: directories.

* speed up test_instruct_pix2pix_checkpointing_checkpoints_total_limit

* speed up: test_controlnet_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* speed up test_controlnet_sdxl

* speed up dreambooth tests

* speed up test_dreambooth_lora_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* speed up test_custom_diffusion_checkpointing_checkpoints_total_limit_removes_multiple_checkpoints

* speed up test_text_to_image_lora_sdxl_text_encoder_checkpointing_checkpoints_total_limit

* speed up # checkpoint-2 should have been deleted

* speed up examples/text_to_image/test_text_to_image.py::TextToImage::test_text_to_image_checkpointing_checkpoints_total_limit

* additional speed ups

* style

* initialize alpha too.

* add: test

* remove config parsing

* store rank

* debug

* remove faulty test

* Fix ProdigyOPT in SDXL Dreambooth script

* style

* style

* fix: unscale fp16 gradient problem

* fix for dreambooth lora sdxl

* make the type-casting conditional.

* 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>