GitLab

change to 2025 licensing for remaining

* Use HF Papers

* Apply style fixes

---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

* store cls instead of an obj

* style

* copy transformer

* copy vae

* copy pipeline

* make fix-copies

* refactor; make original code work with diffusers; test latents for comparison generated with this commit

* move rope into pipeline; remove flash attention; refactor

* begin conversion script

* make style

* refactor attention

* refactor

* refactor final layer

* their mlp -> our feedforward

* make style

* add docs

* refactor layer names

* refactor modulation

* cleanup

* refactor norms

* refactor activations

* refactor single blocks attention

* refactor attention processor

* make style

* cleanup a bit

* refactor double transformer block attention

* update mochi attn proc

* use diffusers attention implementation in all modules; checkpoint for all values matching original

* remove helper functions in vae

* refactor upsample

* refactor causal conv

* refactor resnet

* refactor

* refactor

* refactor

* grad checkpointing

* autoencoder test

* fix scaling factor

* refactor clip

* refactor llama text encoding

* add coauthor
Co-Authored-By: "Gregory D. Hunkins" <greg@ollano.com>

* refactor rope; diff: 0.14990234375; reason and fix: create rope grid on cpu and move to device

Note: The following line diverges from original behaviour. We create the grid on the device, whereas
original implementation creates it on CPU and then moves it to device. This results in numerical
differences in layerwise debugging outputs, but visually it is the same.

* use diffusers timesteps embedding; diff: 0.10205078125

* rename

* convert

* update

* add tests for transformer

* add pipeline tests; text encoder 2 is not optional

* fix attention implementation for torch

* add example

* update docs

* update docs

* apply suggestions from review

* refactor vae

* update

* Apply suggestions from code review
Co-authored-by: hlky <hlky@hlky.ac>

* Update src/diffusers/pipelines/hunyuan_video/pipeline_hunyuan_video.py
Co-authored-by: hlky <hlky@hlky.ac>

* Update src/diffusers/pipelines/hunyuan_video/pipeline_hunyuan_video.py
Co-authored-by: hlky <hlky@hlky.ac>

* make fix-copies

* update

---------
Co-authored-by: "Gregory D. Hunkins" <greg@ollano.com>
Co-authored-by: hlky <hlky@hlky.ac>

* Remove mps workaround for fp16 GELU, which is now supported natively

---------
Co-authored-by: hlky <hlky@hlky.ac>

* update

* udpate

* update transformer

* make style

* fix

* add conversion script

* update

* fix

* update

* fix

* update

* fixes

* make style

* update

* update

* update

* init

* update

* update

* add

* up

* up

* up

* update

* mochi transformer

* remove original implementation

* make style

* update inits

* update conversion script

* docs

* Update src/diffusers/pipelines/mochi/pipeline_mochi.py
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>

* Update src/diffusers/pipelines/mochi/pipeline_mochi.py
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>

* fix docs

* pipeline fixes

* make style

* invert sigmas in scheduler; fix pipeline

* fix pipeline num_frames

* flip proj and gate in swiglu

* make style

* fix

* make style

* fix tests

* latent mean and std fix

* update

* cherry-pick 1069d210e1b9e84a366cdc7a13965626ea258178

* remove additional sigma already handled by flow match scheduler

* fix

* remove hardcoded value

* replace conv1x1 with linear

* Update src/diffusers/pipelines/mochi/pipeline_mochi.py
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>

* framewise decoding and conv_cache

* make style

* Apply suggestions from code review

* mochi vae encoder changes

* rebase correctly

* Update scripts/convert_mochi_to_diffusers.py

* fix tests

* fixes

* make style

* update

* make style

* update

* add framewise and tiled encoding

* make style

* make original vae implementation behaviour the default; note: framewise encoding does not work

* remove framewise encoding implementation due to presence of attn layers

* fight test 1

* fight test 2

---------
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
Co-authored-by: yiyixuxu <yixu310@gmail.com>

* WIP modeling code and pipeline

* add custom attention processor + custom activation + add to init

* correct ProjectionModel forward

* add stable audio to __initèè

* add autoencoder and update pipeline and modeling code

* add half Rope

* add partial rotary v2

* add temporary modfis to scheduler

* add EDM DPM Solver

* remove TODOs

* clean GLU

* remove att.group_norm to attn processor

* revert back src/diffusers/schedulers/scheduling_dpmsolver_multistep.py

* refactor GLU -> SwiGLU

* remove redundant args

* add channel multiples in autoencoder docstrings

* changes in docsrtings and copyright headers

* clean pipeline

* further cleaning

* remove peft and lora and fromoriginalmodel

* Delete src/diffusers/pipelines/stable_audio/diffusers.code-workspace

* make style

* dummy models

* fix copied from

* add fast oobleck tests

* add brownian tree

* oobleck autoencoder slow tests

* remove TODO

* fast stable audio pipeline tests

* add slow tests

* make style

* add first version of docs

* wrap is_torchsde_available to the scheduler

* fix slow test

* test with input waveform

* add input waveform

* remove some todos

* create stableaudio gaussian projection + make style

* add pipeline to toctree

* fix copied from

* make quality

* refactor timestep_features->time_proj

* refactor joint_attention_kwargs->cross_attention_kwargs

* remove forward_chunk

* move StableAudioDitModel to transformers folder

* correct convert + remove partial rotary embed

* apply suggestions from yiyixuxu -> removing attn.kv_heads

* remove temb

* remove cross_attention_kwargs

* further removal of cross_attention_kwargs

* remove text encoder autocast to fp16

* continue removing autocast

* make style

* refactor how text and audio are embedded

* add paper

* update example code

* make style

* unify projection model forward + fix device placement

* make style

* remove fuse qkv

* apply suggestions from review

* Update src/diffusers/pipelines/stable_audio/pipeline_stable_audio.py
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* make style

* smaller models in fast tests

* pass sequential offloading fast tests

* add docs for vae and autoencoder

* make style and update example

* remove useless import

* add cosine scheduler

* dummy classes

* cosine scheduler docs

* better description of scheduler

---------
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* Hunyuan Team: add HunyuanDiT related updates


---------
Co-authored-by: XCLiu <liuxc1996@gmail.com>
Co-authored-by: yiyixuxu <yixu310@gmail.com>

Add Ascend NPU support for SDXL fine-tuning and fix the model saving bug when using DeepSpeed. (#7816)

* Add Ascend NPU support for SDXL fine-tuning and fix the model saving bug when using DeepSpeed.

* fix check code quality

* Decouple the NPU flash attention and make it an independent module.

* add doc and unit tests for npu flash attention.

---------
Co-authored-by: mhh001 <mahonghao1@huawei.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* fix PyTorch classes and start deprecsation cycles.

* remove args crafting for accommodating scale.

* remove scale check in feedforward.

* assert against nn.Linear and not CompatibleLinear.

* remove conv_cls and lineaR_cls.

* remove scale

* 👋

 scale.

* fix: unet2dcondition

* fix attention.py

* fix: attention.py again

* fix: unet_2d_blocks.

* fix-copies.

* more fixes.

* fix: resnet.py

* more fixes

* fix i2vgenxl unet.

* depcrecate scale gently.

* fix-copies

* Apply suggestions from code review
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* quality

* throw warning when scale is passed to the the BasicTransformerBlock class.

* remove scale from signature.

* cross_attention_kwargs, very nice catch by Yiyi

* fix: logger.warn

* make deprecation message clearer.

* address final comments.

* maintain same depcrecation message and also add it to activations.

* address yiyi

* fix copies

* Apply suggestions from code review
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* more depcrecation

* fix-copies

---------
Co-authored-by: YiYi Xu <yixu310@gmail.com>

change to 2024

* Support IP-Adapter Plus

* fix format

* restore before black format

* restore before black format

* generic

* Refactor PerceiverAttention

* format

* fix test and refactor PerceiverAttention

* generic encode_image

* keep attention implementation

* merge tests

* encode_image backward compatible

* code quality

* fix controlnet inpaint pipeline

* refactor FFN

* refactor FFN

---------
Co-authored-by: YiYi Xu <yixu310@gmail.com>

* I added a new doc string to the class. This is more flexible to understanding other developers what are doing and where it's using.

* Update src/diffusers/models/unet_2d_blocks.py

This changes suggest by maintener.
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

* Update src/diffusers/models/unet_2d_blocks.py

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

* Update unet_2d_blocks.py

I changed the Parameter to Args text.

* Update unet_2d_blocks.py

proper indentation set in this file.

* Update unet_2d_blocks.py

a little bit of change in the act_fun argument line.

* I run the black command to reformat style in the code

* Update unet_2d_blocks.py

similar doc-string add to have in the original diffusion repository.

* I use a lower method in the activation function.

* Replace multiple if-else statements with a dictionary of activation functions, and call one if statement to retrieve the appropriate function.

* I am using black package to reforamted my file

* I defined the ACTIVATION_FUNCTIONS variable outside of the function

* activation function variable convert to lower case

* First, I resolved the conflict issue. Then, I ran the Black package to reformat my file.

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

* move out the activations.

* move normalization layers.

* add doc.

* add doc.

* fix: paths

* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* style

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* improvement: add typehints and docs to diffusers/models/activations.py

* improvement: add typehints and docs to diffusers/models/resnet.py

* add: model implementation of tiny autoencoder.

* add: inits.

* push the latest devs.

* add: conversion script and finish.

* add: scaling factor args.

* debugging

* fix denormalization.

* fix: positional argument.

* handle use_torch_2_0_or_xformers.

* handle post_quant_conv

* handle dtype

* fix: sdxl image processor for tiny ae.

* fix: sdxl image processor for tiny ae.

* unify upcasting logic.

* copied from madness.

* remove trailing whitespace.

* set is_tiny_vae = False

* address PR comments.

* change to AutoencoderTiny

* make act_fn an str throughout

* fix: apply_forward_hook decorator call

* get rid of the special is_tiny_vae flag.

* directly scale the output.

* fix dummies?

* fix: act_fn.

* get rid of the Clamp() layer.

* bring back copied from.

* movement of the blocks to appropriate modules.

* add: docstrings to AutoencoderTiny

* add: documentation.

* changes to the conversion script.

* add doc entry.

* settle tests.

* style

* add one slow test.

* fix

* fix 2

* fix 2

* fix: 4

* fix: 5

* finish integration tests

* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* style

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* move activation dispatches into helper function

* tests