GitLab

* feat : add log-rho deis multistep deis

* docs :fix typo

* docs : add docs for impl algo

* docs : remove duplicate ref

* finish deis

* add docs

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

* [Repro] Correct reproducability

* up

* up

* uP

* up

* need better image

* allow conversion from no state dict checkpoints

* up

* up

* up

* up

* check tensors

* check tensors

* check tensors

* check tensors

* next try

* up

* up

* better name

* up

* up

* Apply suggestions from code review

* correct more

* up

* replace all torch randn

* fix

* correct

* correct

* finish

* fix more

* up

* Support training SD V2 with Flax

Mostly involves supporting a v_prediction scheduler.

The implementation in #1777 doesn't take into account a recent refactor of `scheduling_utils_flax`, so this should be used instead.

* Add to other top-level files.

* [Deterministic torch randn] Allow tensors to be generated on CPU

* fix more

* up

* fix more

* up

* Update src/diffusers/utils/torch_utils.py
Co-authored-by: Anton Lozhkov <anton@huggingface.co>

* Apply suggestions from code review

* up

* up

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

* add add_noise to scheduling-sde-ve

* run Black formater

Don't initialize Jax on startup.

* [Flax] Stateless schedulers, fixes and refactors

* Remove scheduling_common_flax and some renames

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

* [wip] attention block updates

* [wip] unCLIP unet decoder and super res

* [wip] unCLIP prior transformer

* [wip] scheduler changes

* [wip] text proj utility class

* [wip] UnCLIPPipeline

* [wip] kakaobrain unCLIP convert script

* [unCLIP pipeline] fixes re: @patrickvonplaten

remove callbacks

move denoising loops into call function

* UNCLIPScheduler re: @patrickvonplaten

Revert changes to DDPMScheduler. Make UNCLIPScheduler, a modified
DDPM scheduler with changes to support karlo

* mask -> attention_mask re: @patrickvonplaten

* [DDPMScheduler] remove leftover change

* [docs] PriorTransformer

* [docs] UNet2DConditionModel and UNet2DModel

* [nit] UNCLIPScheduler -> UnCLIPScheduler

matches existing unclip naming better

* [docs] SchedulingUnCLIP

* [docs] UnCLIPTextProjModel

* refactor

* finish licenses

* rename all to attention_mask and prep in models

* more renaming

* don't expose unused configs

* final renaming fixes

* remove x attn mask when not necessary

* configure kakao script to use new class embedding config

* fix copies

* [tests] UnCLIPScheduler

* finish x attn

* finish

* remove more

* rename condition blocks

* clean more

* Apply suggestions from code review

* up

* fix

* [tests] UnCLIPPipelineFastTests

* remove unused imports

* [tests] UnCLIPPipelineIntegrationTests

* correct

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

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* add fast tests

* better tests and fp16

* batch fixes

* Reuse preprocessing

* quickfix

upgrade version

* Fix PyCharm/VSCode static type checking for dummy objects

* Re-add dummies

* Fix AudioDiffusion imports

* fix import

* fix import

* Update utils/check_dummies.py
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* Update src/diffusers/utils/import_utils.py

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

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

* fix double import
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Update scheduling_repaint.py

* update the expected image
Co-authored-by: anton- <anton@huggingface.co>

* add singlestep dpmsolver

* fix a style typo

* fix a style typo

* add docs

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

- Add the missing `scale_model_input` method to `FlaxLMSDiscreteScheduler`
- Use `jnp.append` for appending to `state.derivatives`
- Use `jnp.delete` to pop from `state.derivatives`

* up

* up

* finish

* finish

* up

* up

* finish

* support v prediction in other schedulers

* v heun

* add tests for v pred

* fix tests

* fix test euler a

* v ddpm

* cast to float for quantile method

* add fp16 test for DPMSolverMultistepScheduler fix

* formatting update

* Add heun

* Finish first version of heun

* remove bogus

* finish

* finish

* improve

* up

* up

* fix more

* change progress bar

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

* finish

* up

* up

* up

* add get_velocity

* add v prediction for training

* fix saving

* add revision arg

* fix saving

* save checkpoints dreambooth

* fix saving embeds

* add instruction in readme

* quality

* noise_pred -> model_pred

* added initial v-pred support to DPM-solver

* fix sign

* added v_prediction to flax

* fixed typo

* fix

* fix deprecated kwargs logic

* add tests

* finish

* Adapt ddpm, ddpmsolver to prediction_type.

* Deprecate predict_epsilon in __init__.

* Bring FlaxDDIMScheduler up to date with DDIMScheduler.

* Set prediction_type as an ivar for consistency.

* Convert pipeline_ddpm

* Adapt tests.

* Adapt unconditional training script.

* Adapt BitDiffusion example.

* Add missing kwargs in dpmsolver_multistep

* Ugly workaround to accept deprecated predict_epsilon when loading
schedulers using from_pretrained.

* make style

* Remove import no longer in use.

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

* Use config.prediction_type everywhere

* Add a couple of Flax prediction type tests.

* make style

* fix register deprecated arg
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* add v prediction

* adat euler for v pred

* velocity -> v_prediction

* simplify

* fix naming

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

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

Casting `self.sigmas` into a different dtype (the one of original_samples) is not advisable. In my img2img pipeline this leads to a long running time in the  `integrate.quad` call later on- by long I mean more than 10x slower.
Co-authored-by: Anton Lozhkov <anton@huggingface.co>

[FLAX] Fix loading scheduler from subfolder

* add conversion script for vae

* uP

* uP

* more changes

* push

* up

* finish again

* up

* up

* up

* up

* finish

* up

* uP

* up

* Apply suggestions from code review
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Anton Lozhkov <anton@huggingface.co>
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* up

* up
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Anton Lozhkov <anton@huggingface.co>
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* re-add RL model code

* match model forward api

* add register_to_config, pass training tests

* fix tests, update forward outputs

* remove unused code, some comments

* add to docs

* remove extra embedding code

* unify time embedding

* remove conv1d output sequential

* remove sequential from conv1dblock

* style and deleting duplicated code

* clean files

* remove unused variables

* clean variables

* add 1d resnet block structure for downsample

* rename as unet1d

* fix renaming

* rename files

* add get_block(...) api

* unify args for model1d like model2d

* minor cleaning

* fix docs

* improve 1d resnet blocks

* fix tests, remove permuts

* fix style

* add output activation

* rename flax blocks file

* Add Value Function and corresponding example script to Diffuser implementation (#884)

* valuefunction code

* start example scripts

* missing imports

* bug fixes and placeholder example script

* add value function scheduler

* load value function from hub and get best actions in example

* very close to working example

* larger batch size for planning

* more tests

* merge unet1d changes

* wandb for debugging, use newer models

* success!

* turns out we just need more diffusion steps

* run on modal

* merge and code cleanup

* use same api for rl model

* fix variance type

* wrong normalization function

* add tests

* style

* style and quality

* edits based on comments

* style and quality

* remove unused var

* hack unet1d into a value function

* add pipeline

* fix arg order

* add pipeline to core library

* community pipeline

* fix couple shape bugs

* style

* Apply suggestions from code review
Co-authored-by: Nathan Lambert <nathan@huggingface.co>

* update post merge of scripts

* add mdiblock / outblock architecture

* Pipeline cleanup (#947)

* valuefunction code

* start example scripts

* missing imports

* bug fixes and placeholder example script

* add value function scheduler

* load value function from hub and get best actions in example

* very close to working example

* larger batch size for planning

* more tests

* merge unet1d changes

* wandb for debugging, use newer models

* success!

* turns out we just need more diffusion steps

* run on modal

* merge and code cleanup

* use same api for rl model

* fix variance type

* wrong normalization function

* add tests

* style

* style and quality

* edits based on comments

* style and quality

* remove unused var

* hack unet1d into a value function

* add pipeline

* fix arg order

* add pipeline to core library

* community pipeline

* fix couple shape bugs

* style

* Apply suggestions from code review

* clean up comments

* convert older script to using pipeline and add readme

* rename scripts

* style, update tests

* delete unet rl model file

* remove imports in src
Co-authored-by: Nathan Lambert <nathan@huggingface.co>

* Update src/diffusers/models/unet_1d_blocks.py

* Update tests/test_models_unet.py

* RL Cleanup v2 (#965)

* valuefunction code

* start example scripts

* missing imports

* bug fixes and placeholder example script

* add value function scheduler

* load value function from hub and get best actions in example

* very close to working example

* larger batch size for planning

* more tests

* merge unet1d changes

* wandb for debugging, use newer models

* success!

* turns out we just need more diffusion steps

* run on modal

* merge and code cleanup

* use same api for rl model

* fix variance type

* wrong normalization function

* add tests

* style

* style and quality

* edits based on comments

* style and quality

* remove unused var

* hack unet1d into a value function

* add pipeline

* fix arg order

* add pipeline to core library

* community pipeline

* fix couple shape bugs

* style

* Apply suggestions from code review

* clean up comments

* convert older script to using pipeline and add readme

* rename scripts

* style, update tests

* delete unet rl model file

* remove imports in src

* add specific vf block and update tests

* style

* Update tests/test_models_unet.py
Co-authored-by: Nathan Lambert <nathan@huggingface.co>

* fix quality in tests

* fix quality style, split test file

* fix checks / tests

* make timesteps closer to main

* unify block API

* unify forward api

* delete lines in examples

* style

* examples style

* all tests pass

* make style

* make dance_diff test pass

* Refactoring RL PR (#1200)

* init file changes

* add import utils

* finish cleaning files, imports

* remove import flags

* clean examples

* fix imports, tests for merge

* update readmes

* hotfix for tests

* quality

* fix some tests

* change defaults

* more mps test fixes

* unet1d defaults

* do not default import experimental

* defaults for tests

* fix tests

* fix-copies

* fix

* changes per Patrik's comments (#1285)

* changes per Patrik's comments

* update conversion script

* fix renaming

* skip more mps tests

* last test fix

* Update examples/rl/README.md
Co-authored-by: Ben Glickenhaus <benglickenhaus@gmail.com>

* Match the generator device to the pipeline for DDPM and DDIM

* style

* fix

* update values

* fix fast tests

* trigger slow tests

* deprecate

* last value fixes

* mps fixes

* fix tests

* Fix more

* more

* [Scheduler] Move predict epsilon to init

* up

* uP

* uP

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

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

* Schedulers: don't use float64 on mps

* Test set_timesteps() on device (float schedulers).

* SD pipeline: use device in set_timesteps.

* SD in-painting pipeline: use device in set_timesteps.

* Tests: fix mps crashes.

* Skip test_load_pipeline_from_git on mps.

Not compatible with float16.

* Use device.type instead of str in Euler schedulers.