GitLab

* first proposal

* rename

* up

* Apply suggestions from code review

* better

* up

* finish

* up

* rename

* correct versatile

* up

* up

* up

* up

* fix

* Apply suggestions from code review

* make style

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

* add error message
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>

* Make cross-attention check more robust.

* Fix copies.

upcast attention

* make attn slice recursive

* remove set_attention_slice from blocks

* fix copies

* make enable_attention_slicing base class method of DiffusionPipeline

* fix set_attention_slice

* fix set_attention_slice

* fix copies

* add tests

* up

* up

* up

* update

* up

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

* Do not use torch.long in mps

Addresses #1056.

* Use torch.int instead of float.

* Propagate changes.

* Do not silently change float -> int.

* Propagate changes.

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

* Moving the mem efficiient attention activation to the top + recursive

* black, too bad there's no pre-commit ?
Co-authored-by: Benjamin Lefaudeux <benjamin@photoroom.com>

* Support SD2 attention slicing

* Support SD2 attention slicing

* Add more copies

* Use attn_num_head_channels in blocks

* fix-copies

* Update tests

* fix imports

* allow disabling self attention

* add class_embedding

* fix copies

* fix condition

* fix copies

* do_self_attention -> only_cross_attention

* fix copies

* num_classes -> num_class_embeds

* fix default value

* boom boom

* remove duplicate arg

* add use_linear_proj arg

* fix copies

* style

* add fast tests

* use_linear_proj -> use_linear_projection

* up

* convert dual unet

* revert dual attn

* adapt for vd-official

* test the full pipeline

* mixed inference

* mixed inference for text2img

* add image prompting

* fix clip norm

* split text2img and img2img

* fix format

* refactor text2img

* mega pipeline

* add optimus

* refactor image var

* wip text_unet

* text unet end to end

* update tests

* reshape

* fix image to text

* add some first docs

* dual guided pipeline

* fix token ratio

* propose change

* dual transformer as a native module

* DualTransformer(nn.Module)

* DualTransformer(nn.Module)

* correct unconditional image

* save-load with mega pipeline

* remove image to text

* up

* uP

* fix

* up

* final fix

* remove_unused_weights

* test updates

* save progress

* uP

* fix dual prompts

* some fixes

* finish

* style

* finish renaming

* up

* fix

* fix

* fix

* finish
Co-authored-by: anton-l <anton@huggingface.co>

* fix non square images with UNet2DModel and DDIM/DDPM pipelines

* fix unet_2d `sample_size` docstring

* update pipeline tests for unet uncond
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* doc string args shape fix

* fix styling

* Fix documentation typo

* Fix other typo

* 2x speedup using memory efficient attention

* remove einops dependency

* Swap K, M in op instantiation

* Simplify code, remove unnecessary maybe_init call and function, remove unused self.scale parameter

* make xformers a soft dependency

* remove one-liner functions

* change one letter variable to appropriate names

* Remove Env variable dependency, remove MemoryEfficientCrossAttention class and use enable_xformers_memory_efficient_attention method

* Add memory efficient attention toggle to img2img and inpaint pipelines

* Clearer management of xformers' availability

* update optimizations markdown to add info about memory efficient attention

* add benchmarks for TITAN RTX

* More detailed explanation of how the mem eff benchmark were ran

* Removing autocast from optimization markdown

* import_utils: import torch only if is available
Co-authored-by: Nouamane Tazi <nouamane98@gmail.com>

* start

* add more logic

* Update src/diffusers/models/unet_2d_condition_flax.py

* match weights

* up

* make model work

* making class more general, fixing missed file rename

* small fix

* make new conversion work

* up

* finalize conversion

* up

* first batch of variable renamings

* remove c and c_prev var names

* add mid and out block structure

* add pipeline

* up

* finish conversion

* finish

* upload

* more fixes

* Apply suggestions from code review

* add attr

* up

* uP

* up

* finish tests

* finish

* uP

* finish

* fix test

* up

* naming consistency in tests

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

* remove hardcoded 16

* Remove bogus

* fix some stuff

* finish

* improve logging

* docs

* upload
Co-authored-by: Nathan Lambert <nol@berkeley.edu>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Nathan Lambert <nathan@huggingface.co>
Co-authored-by: Anton Lozhkov <anton@huggingface.co>

* add or fix license formatting

* fix quality

fix typo docstring

* Removing `autocast` for `35-25% speedup`.

* iQuality

* Adding a slow test.

* Fixing mps noise generation.

* Raising error on wrong device, instead of just casting on behalf of user.

* Quality.

* fix merge
Co-authored-by: Nouamane Tazi <nouamane98@gmail.com>

Remove comments no longer appropriate.

There were casting operations before, they are now gone.

* Allow resolutions that are not multiples of 64

* ran black

* fix bug

* add test

* more explanation

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

* initial commit

* make UNet stream capturable

* try to fix noise_pred value

* remove cuda graph and keep NB

* non blocking unet with PNDMScheduler

* make timesteps np arrays for pndm scheduler
because lists don't get formatted to tensors in `self.set_format`

* make max async in pndm

* use channel last format in unet

* avoid moving timesteps device in each unet call

* avoid memcpy op in `get_timestep_embedding`

* add `channels_last` kwarg to `DiffusionPipeline.from_pretrained`

* update TODO

* replace `channels_last` kwarg with `memory_format` for more generality

* revert the channels_last changes to leave it for another PR

* remove non_blocking when moving input ids to device

* remove blocking from all .to() operations at beginning of pipeline

* fix merging

* fix merging

* model can run in other precisions without autocast

* attn refactoring

* Revert "attn refactoring"

This reverts commit 0c70c0e189cd2c4d8768274c9fcf5b940ee310fb.

* remove restriction to run conv_norm in fp32

* use `baddbmm` instead of `matmul`for better in attention for better perf

* removing all reshapes to test perf

* Revert "removing all reshapes to test perf"

This reverts commit 006ccb8a8c6bc7eb7e512392e692a29d9b1553cd.

* add shapes comments

* hardcore whats needed for jitting

* Revert "hardcore whats needed for jitting"

This reverts commit 2fa9c698eae2890ac5f8e367ca80532ecf94df9a.

* Revert "remove restriction to run conv_norm in fp32"

This reverts commit cec592890c32da3d1b78d38b49e4307aedf459b9.

* revert using baddmm in attention's forward

* cleanup comment

* remove restriction to run conv_norm in fp32. no quality loss was noticed

This reverts commit cc9bc1339c998ebe9e7d733f910c6d72d9792213.

* add more optimizations techniques to docs

* Revert "add shapes comments"

This reverts commit 31c58eadb8892f95478cdf05229adf678678c5f4.

* apply suggestions

* make quality

* apply suggestions

* styling

* `scheduler.timesteps` are now arrays so we dont need .to()

* remove useless .type()

* use mean instead of max in `test_stable_diffusion_inpaint_pipeline_k_lms`

* move scheduler timestamps to correct device if tensors

* add device to `set_timesteps` in LMSD scheduler

* `self.scheduler.set_timesteps` now uses device arg for schedulers that accept it

* quick fix

* styling

* remove kwargs from schedulers `set_timesteps`

* revert to using max in K-LMS inpaint pipeline test

* Revert "`self.scheduler.set_timesteps` now uses device arg for schedulers that accept it"

This reverts commit 00d5a51e5c20d8d445c8664407ef29608106d899.

* move timesteps to correct device before loop in SD pipeline

* apply previous fix to other SD pipelines

* UNet now accepts tensor timesteps even on wrong device, to avoid errors
- it shouldnt affect performance if timesteps are alrdy on correct device
- it does slow down performance if they're on the wrong device

* fix pipeline when timesteps are arrays with strides

* add grad ckpt to downsample blocks

* make it work

* don't pass gradient_checkpointing to upsample block

* add tests for UNet2DConditionModel

* add test_gradient_checkpointing

* add gradient_checkpointing for up and down blocks

* add functions to enable and disable grad ckpt

* remove the forward argument

* better naming

* make supports_gradient_checkpointing private

* First UNet Flax modeling blocks.

Mimic the structure of the PyTorch files.
The model classes themselves need work, depending on what we do about
configuration and initialization.

* Remove FlaxUNet2DConfig class.

* ignore_for_config non-config args.

* Implement `FlaxModelMixin`

* Use new mixins for Flax UNet.

For some reason the configuration is not correctly applied; the
signature of the `__init__` method does not contain all the parameters
by the time it's inspected in `extract_init_dict`.

* Import `FlaxUNet2DConditionModel` if flax is available.

* Rm unused method `framework`

* Update src/diffusers/modeling_flax_utils.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Indicate types in flax.struct.dataclass as pointed out by @mishig25
Co-authored-by: Mishig Davaadorj <mishig.davaadorj@coloradocollege.edu>

* Fix typo in transformer block.

* make style

* some more changes

* make style

* Add comment

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

* Rm unneeded comment

* Update docstrings

* correct ignore kwargs

* make style

* Update docstring examples

* Make style

* Style: remove empty line.

* Apply style (after upgrading black from pinned version)

* Remove some commented code and unused imports.

* Add init_weights (not yet in use until #513).

* Trickle down deterministic to blocks.

* Rename q, k, v according to the latest PyTorch version.

Note that weights were exported with the old names, so we need to be
careful.

* Flax UNet docstrings, default props as in PyTorch.

* Fix minor typos in PyTorch docstrings.

* Use FlaxUNet2DConditionOutput as output from UNet.

* make style
Co-authored-by: Mishig Davaadorj <dmishig@gmail.com>
Co-authored-by: Mishig Davaadorj <mishig.davaadorj@coloradocollege.edu>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* pass norm_num_groups to unet blocs and attention

* fix UNet2DConditionModel

* add norm_num_groups arg in vae

* add tests

* remove comment

* Apply suggestions from code review

* docs for attention

* types for embeddings

* unet2d docstrings

* UNet2DConditionModel docstrings

* fix typos

* style and vq-vae docstrings

* docstrings  for VAE

* Update src/diffusers/models/unet_2d.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* make style

* added inherits from sentence

* docstring to forward

* make style

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

* finish model docs

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

* Initial support for mps in Stable Diffusion pipeline.

* Initial "warmup" implementation when using mps.

* Make some deterministic tests pass with mps.

* Disable training tests when using mps.

* SD: generate latents in CPU then move to device.

This is especially important when using the mps device, because
generators are not supported there. See for example
https://github.com/pytorch/pytorch/issues/84288.

In addition, the other pipelines seem to use the same approach: generate
the random samples then move to the appropriate device.

After this change, generating an image in MPS produces the same result
as when using the CPU, if the same seed is used.

* Remove prints.

* Pass AutoencoderKL test_output_pretrained with mps.

Sampling from `posterior` must be done in CPU.

* Style

* Do not use torch.long for log op in mps device.

* Perform incompatible padding ops in CPU.

UNet tests now pass.
See https://github.com/pytorch/pytorch/issues/84535



* Style: fix import order.

* Remove unused symbols.

* Remove MPSWarmupMixin, do not apply automatically.

We do apply warmup in the tests, but not during normal use.
This adopts some PR suggestions by @patrickvonplaten.

* Add comment for mps fallback to CPU step.

* Add README_mps.md for mps installation and use.

* Apply `black` to modified files.

* Restrict README_mps to SD, show measures in table.

* Make PNDM indexing compatible with mps.

Addresses #239.

* Do not use float64 when using LDMScheduler.

Fixes #358.

* Fix typo identified by @patil-suraj
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Adapt example to new output style.

* Restore 1:1 results reproducibility with CompVis.

However, mps latents need to be generated in CPU because generators
don't work in the mps device.

* Move PyTorch nightly to requirements.

* Adapt `test_scheduler_outputs_equivalence` ton MPS.

* mps: skip training tests instead of ignoring silently.

* Make VQModel tests pass on mps.

* mps ddim tests: warmup, increase tolerance.

* ScoreSdeVeScheduler indexing made mps compatible.

* Make ldm pipeline tests pass using warmup.

* Style

* Simplify casting as suggested in PR.

* Add Known Issues to readme.

* `isort` import order.

* Remove _mps_warmup helpers from ModelMixin.

And just make changes to the tests.

* Skip tests using unittest decorator for consistency.

* Remove temporary var.

* Remove spurious blank space.

* Remove unused symbol.

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

* up

* add tests

* correct

* up

* finish

* better naming

* Update README.md
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

Use `expand` instead of ones to broadcast tensor.

As suggested by @bes-dev. According the documentation this shouldn't
take any memory - it just plays with the strides.

* add outputs for models

* add for pipelines

* finish schedulers

* better naming

* adapt tests as well

* replace dict access with . access

* make schedulers works

* finish

* correct readme

* make  bcp compatible

* up

* small fix

* finish

* more fixes

* more fixes

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

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

* Adapt model outputs

* Apply more suggestions

* finish examples

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

* add void check

* remove void, add types for params

* Use ONNX / Core ML compatible method to broadcast.

Unfortunately `tile` could not be used either, it's still not compatible
with ONNX.

See #284.

* Add comment about why broadcast_to is not used.

Also, apply style to changed files.

* Make sure broadcast remains in same device.

* [Half precision] Make sure half-precision is correct

* Update src/diffusers/models/unet_2d.py

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

* correct some tests

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

* finalize

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

add cross_attention_dim as an argument

* up

* change model name

* renaming

* more changes

* up

* up

* up

* save checkpoint

* finish api / naming

* finish config renaming

* rename all weights

* finish really