GitLab

* 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>

* 0.20.0dev0

* make style

* fix circular import

* fix imports when watermark not specified

* fix all pipelines

* 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>

* up

* finish

* fix final

* Allow vae to be loaded

* up

* 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>

fix: missed network_alpha when loading lora from old format

* 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

* Add controlnet from single file

* Updates

* make style

* finish

* Apply suggestions from code review
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

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

replace with os path join

* add: test for testing unloading lora.

* add :reason to skipif.

* initial implementation of lora unload().

* apply styling.

* add: doc.

* change checkpoints.

* reinit generator

* finalize slow test.

* add fast test for unloading lora.

* start improving single file load

* Fix more

* start improving single file load

* Fix sd 2.1

* further improve from_single_file

* refactor to support patching LoRA into T5

instantiate the lora linear layer on the same device as the regular linear layer

get lora rank from state dict

tests

fmt

can create lora layer in float32 even when rest of model is float16

fix loading model hook

remove load_lora_weights_ and T5 dispatching

remove Unet#attn_processors_state_dict

docstrings

* text encoder monkeypatch class method

* fix test

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

* Update pipeline_utils.py

typo in safetensors (safetenstors)

* Update loaders.py

typo in safetensors (safetenstors)

* Update modeling_utils.py

typo in safetensors (safetenstors)

* improve sd xl

* correct more

* finish

* make style

* fix more

* Support for manual CLIP loading in StableDiffusionPipeline - txt2img.

* Update src/diffusers/pipelines/stable_diffusion/convert_from_ckpt.py

* Update variables & according docs to match previous style.

* Updated to match style & quality of 'diffusers'

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

* clean up loaders

* clean up rest of main class apis

* apply feedback

* Post release

* Post release

* Add draft for lora text encoder scale

* Improve naming

* fix: training dreambooth lora script.

* Apply suggestions from code review

* Update examples/dreambooth/train_dreambooth_lora.py

* Apply suggestions from code review

* Apply suggestions from code review

* add lora mixin when fit

* add lora mixin when fit

* add lora mixin when fit

* fix more

* fix more

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

* feat: add lora attention processor for pt 2.0.

* explicit context manager for SDPA.

* switch to flash attention

* make shapes compatible to work optimally with SDPA.

* fix: circular import problem.

* explicitly specify the flash attention kernel in sdpa

* fall back to efficient attention context manager.

* remove explicit dispatch.

* fix: removed processor.

* fix: remove optional from type annotation.

* feat: make changes regarding LoRAAttnProcessor2_0.

* remove confusing warning.

* formatting.

* relax tolerance for PT 2.0

* fix: loading message.

* remove unnecessary logging.

* add: entry to the docs.

* add: network_alpha argument.

* relax tolerance.

* merge undoable-monkeypatch

* remove TEXT_ENCODER_TARGET_MODULES, refactoring

* move create_lora_weight_file

* iterate over unique tokens to avoid duplicate replacements

* added test for multiple references to multi embedding

* adhere to black formatting

* reorder test post-rebase

* add _convert_kohya_lora_to_diffusers

* make style

* add scaffold

* match result: unet attention only

* fix monkey-patch for text_encoder

* with CLIPAttention

While the terrible images are no longer produced,
the results do not match those from the hook ver.
This may be due to not setting the network_alpha value.

* add to support network_alpha

* generate diff image

* fix monkey-patch for text_encoder

* add test_text_encoder_lora_monkey_patch()

* verify that it's okay to release the attn_procs

* fix closure version

* add comment

* Revert "fix monkey-patch for text_encoder"

This reverts commit bb9c61e6faecc1935c9c4319c77065837655d616.

* Fix to reuse utility functions

* make LoRAAttnProcessor targets to self_attn

* fix LoRAAttnProcessor target

* make style

* fix split key

* Update src/diffusers/loaders.py

* remove TEXT_ENCODER_TARGET_MODULES loop

* add print memory usage

* remove test_kohya_loras_scaffold.py

* add: doc on LoRA civitai

* remove print statement and refactor in the doc.

* fix state_dict test for kohya-ss style lora

* Apply suggestions from code review
Co-authored-by: Takuma Mori <takuma104@gmail.com>

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

* enable state dict for textual inversion loader

* Empty-Commit | restart CI

* Empty-Commit | restart CI

* Empty-Commit | restart CI

* Empty-Commit | restart CI

* add tests

* fix tests

* fix tests

* fix tests

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

Fix to apply LoRAXFormersAttnProcessor instead of LoRAAttnProcessor when xFormers is enabled (#3556)

* fix to use LoRAXFormersAttnProcessor

* add test

* using new LoraLoaderMixin.save_lora_weights

* add test_lora_save_load_with_xformers

* Correct from_ckpt

* make style

* Improve checkpointing lora

* fix more

* Improve doc string

* Update src/diffusers/loaders.py

* make stytle

* Apply suggestions from code review

* Update src/diffusers/loaders.py

* Apply suggestions from code review

* Apply suggestions from code review

* better

* Fix all

* Fix multi-GPU dreambooth

* Apply suggestions from code review
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* Fix all

* make style

* make style

---------
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* clarify safetensor docstring

* fix typo

* apply feedback

* update IF stage I pipelines

add fixed variance schedulers and lora loading

* added kv lora attn processor

* allow loading into alternative lora attn processor

* make vae optional

* throw away predicted variance

* allow loading into added kv lora layer

* allow load T5

* allow pre compute text embeddings

* set new variance type in schedulers

* fix copies

* refactor all prompt embedding code

class prompts are now included in pre-encoding code
max tokenizer length is now configurable
embedding attention mask is now configurable

* fix for when variance type is not defined on scheduler

* do not pre compute validation prompt if not present

* add example test for if lora dreambooth

* add check for train text encoder and pre compute text embeddings

* Batched load of textual inversions

- Only call resize_token_embeddings once per batch as it is the most expensive operation
- Allow pretrained_model_name_or_path and token to be an optional list
- Remove Dict from type annotation pretrained_model_name_or_path as it was not supported in this function
- Add comment that single files (e.g. .pt/.safetensors) are supported
- Add comment for token parameter
- Convert token override log message from warning to info

* Update src/diffusers/loaders.py

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

* Update condition for None tokens

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

* 👽

 qol improvements for LoRA.

* better function name?

* fix: LoRA weight loading with the new format.

* address Patrick's comments.

* Apply suggestions from code review
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* change wording around encouraging the use of load_lora_weights().

* fix: function name.

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

* Fix docs text inversion

* Apply suggestions from code review

When the token used for textual inversion does not have any special symbols (e.g. it is not surrounded by <>), the tokenizer does not properly split the replacement tokens.  Adding a space for the padding tokens fixes this.