GitLab

[Variant] Add "variant" as input kwarg so to have better UX when downloading no_ema or fp16 weights (#2305)

* [Variant] Add variant loading mechanism

* clean

* improve further

* up

* add tests

* add some first tests

* up

* up

* use path splittetx

* add deprecate

* deprecation warnings

* improve docs

* up

* up

* up

* fix tests

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

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

* correct code format

* fix warning

* finish

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

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

* Update docs/source/en/using-diffusers/loading.mdx
Co-authored-by: Suraj Patil <surajp815@gmail.com>

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

* correct loading docs

* finish

---------
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Will Berman <wlbberman@gmail.com>

correctly locate 3rd file; also correct misleading docs

* add total number checkpoints to training scripts

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

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

There isn't a space between the "Scope" paragraph and "Ethical Guidelines", here: https://huggingface.co/docs/diffusers/main/en/conceptual/ethical_guidelines , yet I can't see that in the preview. In this PR, I'm simply adding some spaces in the hopes that it resolves the issue.....

* finish

* Apply suggestions from code review

* fix indent and import error in test_stable_diffusion_depth

---------
Co-authored-by: William Berman <WLBberman@gmail.com>

* initial docs about KarrasDiffusionSchedulers

* typo

* grammer

* Update docs/source/en/api/schedulers/overview.mdx
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* do not list the schedulers explicitly

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

replace type hints

* Log Unconditional Image Generation Samples to WandB

* Check for wandb installation and parity between onnxruntime script

* Log epoch to wandb

* Check for tensorboard logger early on

* style fixes

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

* pipeline_variant

* Add docs for when clip_stats_path is specified

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_unclip.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_unclip.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_unclip_img2img.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update src/diffusers/pipelines/stable_diffusion/pipeline_stable_unclip_img2img.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* prepare_latents # Copied from re: @patrickvonplaten

* NoiseAugmentor->ImageNormalizer

* stable_unclip_prior default to None re: @patrickvonplaten

* prepare_prior_extra_step_kwargs

* prior denoising scale model input

* {DDIM,DDPM}Scheduler -> KarrasDiffusionSchedulers re: @patrickvonplaten

* docs

* Update docs/source/en/api/pipelines/stable_unclip.mdx
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

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

* unet check length input

* prep test file for changes

* correct all tests

* clean up

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

* github issue forum link

* Update .github/ISSUE_TEMPLATE/config.yml
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

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

fix typo

* Fix running LoRA with xformers

* support disabling xformers

* reformat

* Add test

* add ethical guidelines

* update file name

* edit file name

* update toctree

* Update docs/source/en/conceptual/ethical_guidelines.mdx

* Update docs/source/en/conceptual/ethical_guidelines.mdx

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

* [Tests] Refactor push tests

* correct

* UnCLIP Text Interpolation Pipeline

* Formatter fixes

* Changes based on feedback

* Formatting fix

* Formatting fix

* isort formatting fix(?)

* Remove duplicate code

* Formatting fix

* Refactor __call__ and change example in readme.

* Update examples/community/unclip_text_interpolation.py

Refactor to linter formatting
Co-authored-by: Will Berman <wlbberman@gmail.com>

---------
Co-authored-by: Will Berman <wlbberman@gmail.com>

* convert ckpt script docstring fixes

* Update src/diffusers/pipelines/stable_diffusion/convert_from_ckpt.py
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* Update src/diffusers/pipelines/stable_diffusion/convert_from_ckpt.py
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

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

* remove ddpm test_full_inference

* style

* correct some

* Apply suggestions from code review

* correct

* Update tests/pipelines/altdiffusion/test_alt_diffusion_img2img.py

* Final

add fast tests

* [LoRA] Freezing the model weights

Freeze the model weights since we don't need to calculate grads for them.

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

* Apply suggestions from code review

---------
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Suraj Patil <surajp815@gmail.com>

Resolves ValueError: `num_inference_steps`: 1000 cannot be larger than `self.config.train_timesteps`: 50 as the unet model trained with this scheduler can only handle maximal 50 timesteps.

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

* [Examples] Test all examples on CPU

* add

* correct

* Apply suggestions from code review

* EMA: fix `state_dict()` & add `cur_decay_value`

* EMA: fix a bug in `load_state_dict()`

'float' object (`state_dict["power"]`) has no attribute 'get'.

* del train_unconditional_ort.py

* Quality check and adding tokenizer

* Adapted stable diffusion to mixed precision+finished up style fixes

* Fixed based on patrick's review

* Fixed oom from number of validation images

* Removed unnecessary np.array conversion

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