* [WIP][LoRA] Implement hot-swapping of LoRA
This PR adds the possibility to hot-swap LoRA adapters. It is WIP.
Description
As of now, users can already load multiple LoRA adapters. They can
offload existing adapters or they can unload them (i.e. delete them).
However, they cannot "hotswap" adapters yet, i.e. substitute the weights
from one LoRA adapter with the weights of another, without the need to
create a separate LoRA adapter.
Generally, hot-swapping may not appear not super useful but when the
model is compiled, it is necessary to prevent recompilation. See #9279
for more context.
Caveats
To hot-swap a LoRA adapter for another, these two adapters should target
exactly the same layers and the "hyper-parameters" of the two adapters
should be identical. For instance, the LoRA alpha has to be the same:
Given that we keep the alpha from the first adapter, the LoRA scaling
would be incorrect for the second adapter otherwise.
Theoretically, we could override the scaling dict with the alpha values
derived from the second adapter's config, but changing the dict will
trigger a guard for recompilation, defeating the main purpose of the
feature.
I also found that compilation flags can have an impact on whether this
works or not. E.g. when passing "reduce-overhead", there will be errors
of the type:
> input name: arg861_1. data pointer changed from 139647332027392 to
139647331054592
I don't know enough about compilation to determine whether this is
problematic or not.
Current state
This is obviously WIP right now to collect feedback and discuss which
direction to take this. If this PR turns out to be useful, the
hot-swapping functions will be added to PEFT itself and can be imported
here (or there is a separate copy in diffusers to avoid the need for a
min PEFT version to use this feature).
Moreover, more tests need to be added to better cover this feature,
although we don't necessarily need tests for the hot-swapping
functionality itself, since those tests will be added to PEFT.
Furthermore, as of now, this is only implemented for the unet. Other
pipeline components have yet to implement this feature.
Finally, it should be properly documented.
I would like to collect feedback on the current state of the PR before
putting more time into finalizing it.
* Reviewer feedback
* Reviewer feedback, adjust test
* Fix, doc
* Make fix
* Fix for possible g++ error
* Add test for recompilation w/o hotswapping
* Make hotswap work
Requires https://github.com/huggingface/peft/pull/2366
More changes to make hotswapping work. Together with the mentioned PEFT
PR, the tests pass for me locally.
List of changes:
- docstring for hotswap
- remove code copied from PEFT, import from PEFT now
- adjustments to PeftAdapterMixin.load_lora_adapter (unfortunately, some
state dict renaming was necessary, LMK if there is a better solution)
- adjustments to UNet2DConditionLoadersMixin._process_lora: LMK if this
is even necessary or not, I'm unsure what the overall relationship is
between this and PeftAdapterMixin.load_lora_adapter
- also in UNet2DConditionLoadersMixin._process_lora, I saw that there is
no LoRA unloading when loading the adapter fails, so I added it
there (in line with what happens in PeftAdapterMixin.load_lora_adapter)
- rewritten tests to avoid shelling out, make the test more precise by
making sure that the outputs align, parametrize it
- also checked the pipeline code mentioned in this comment:
https://github.com/huggingface/diffusers/pull/9453#issuecomment-2418508871;
when running this inside the with
torch._dynamo.config.patch(error_on_recompile=True) context, there is
no error, so I think hotswapping is now working with pipelines.
* Address reviewer feedback:
- Revert deprecated method
- Fix PEFT doc link to main
- Don't use private function
- Clarify magic numbers
- Add pipeline test
Moreover:
- Extend docstrings
- Extend existing test for outputs != 0
- Extend existing test for wrong adapter name
* Change order of test decorators
parameterized.expand seems to ignore skip decorators if added in last
place (i.e. innermost decorator).
* Split model and pipeline tests
Also increase test coverage by also targeting conv2d layers (support of
which was added recently on the PEFT PR).
* Reviewer feedback: Move decorator to test classes
... instead of having them on each test method.
* Apply suggestions from code review
Co-authored-by: hlky <hlky@hlky.ac>
* Reviewer feedback: version check, TODO comment
* Add enable_lora_hotswap method
* Reviewer feedback: check _lora_loadable_modules
* Revert changes in unet.py
* Add possibility to ignore enabled at wrong time
* Fix docstrings
* Log possible PEFT error, test
* Raise helpful error if hotswap not supported
I.e. for the text encoder
* Formatting
* More linter
* More ruff
* Doc-builder complaint
* Update docstring:
- mention no text encoder support yet
- make it clear that LoRA is meant
- mention that same adapter name should be passed
* Fix error in docstring
* Update more methods with hotswap argument
- SDXL
- SD3
- Flux
No changes were made to load_lora_into_transformer.
* Add hotswap argument to load_lora_into_transformer
For SD3 and Flux. Use shorter docstring for brevity.
* Extend docstrings
* Add version guards to tests
* Formatting
* Fix LoRA loading call to add prefix=None
See:
https://github.com/huggingface/diffusers/pull/10187#issuecomment-2717571064
* Run make fix-copies
* Add hot swap documentation to the docs
* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
Co-authored-by: hlky <hlky@hlky.ac>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
* update
* fix
* non_blocking; handle parameters and buffers
* update
* Group offloading with cuda stream prefetching (#10516)
* cuda stream prefetch
* remove breakpoints
* update
* copy model hook implementation from pab
* update; ~very workaround based implementation but it seems to work as expected; needs cleanup and rewrite
* more workarounds to make it actually work
* cleanup
* rewrite
* update
* make sure to sync current stream before overwriting with pinned params
not doing so will lead to erroneous computations on the GPU and cause bad results
* better check
* update
* remove hook implementation to not deal with merge conflict
* re-add hook changes
* why use more memory when less memory do trick
* why still use slightly more memory when less memory do trick
* optimise
* add model tests
* add pipeline tests
* update docs
* add layernorm and groupnorm
* address review comments
* improve tests; add docs
* improve docs
* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
* apply suggestions from code review
* update tests
* apply suggestions from review
* enable_group_offloading -> enable_group_offload for naming consistency
* raise errors if multiple offloading strategies used; add relevant tests
* handle .to() when group offload applied
* refactor some repeated code
* remove unintentional change from merge conflict
* handle .cuda()
---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
* update
* update
* make style
* remove dynamo disable
* add coauthor
Co-Authored-By: Dhruv Nair <dhruv.nair@gmail.com>
* update
* update
* update
* update mixin
* add some basic tests
* update
* update
* non_blocking
* improvements
* update
* norm.* -> norm
* apply suggestions from review
* add example
* update hook implementation to the latest changes from pyramid attention broadcast
* deinitialize should raise an error
* update doc page
* Apply suggestions from code review
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
* update docs
* update
* refactor
* fix _always_upcast_modules for asym ae and vq_model
* fix lumina embedding forward to not depend on weight dtype
* refactor tests
* add simple lora inference tests
* _always_upcast_modules -> _precision_sensitive_module_patterns
* remove todo comments about review; revert changes to self.dtype in unets because .dtype on ModelMixin should be able to handle fp8 weight case
* check layer dtypes in lora test
* fix UNet1DModelTests::test_layerwise_upcasting_inference
* _precision_sensitive_module_patterns -> _skip_layerwise_casting_patterns based on feedback
* skip test in NCSNppModelTests
* skip tests for AutoencoderTinyTests
* skip tests for AutoencoderOobleckTests
* skip tests for UNet1DModelTests - unsupported pytorch operations
* layerwise_upcasting -> layerwise_casting
* skip tests for UNetRLModelTests; needs next pytorch release for currently unimplemented operation support
* add layerwise fp8 pipeline test
* use xfail
* Apply suggestions from code review
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
* add assertion with fp32 comparison; add tolerance to fp8-fp32 vs fp32-fp32 comparison (required for a few models' test to pass)
* add note about memory consumption on tesla CI runner for failing test
---------
Co-authored-by: Dhruv Nair <dhruv.nair@gmail.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
* fix the Positinoal Embedding bug in 2K model;
* Change the default model to the BF16 one for more stable training and output
* make style
* substract buffer size
* add compute_module_persistent_sizes
---------
Co-authored-by: yiyixuxu <yixu310@gmail.com>
* remove 2 shapes from SDFunctionTesterMixin::test_vae_tiling
* combine freeu enable/disable test to reduce many inference runs
* remove low signal unet test for signature
* remove low signal embeddings test
* remove low signal progress bar test from PipelineTesterMixin
* combine ip-adapter single and multi tests to save many inferences
* fix broken tests
* Update tests/pipelines/test_pipelines_common.py
* Update tests/pipelines/test_pipelines_common.py
* add progress bar tests
* Fix sharding when no device_map is passed
* style
* add tests
* align
* add docstring
* format
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>
* feat: support saving a model in sharded checkpoints.
* feat: make loading of sharded checkpoints work.
* add tests
* cleanse the loading logic a bit more.
* more resilience while loading from the Hub.
* parallelize shard downloads by using snapshot_download()/
* default to a shard size.
* more fix
* Empty-Commit
* debug
* fix
* uality
* more debugging
* fix more
* initial comments from Benjamin
* move certain methods to loading_utils
* add test to check if the correct number of shards are present.
* add a test to check if loading of sharded checkpoints from the Hub is okay
* clarify the unit when passed as an int.
* use hf_hub for sharding.
* remove unnecessary code
* remove unnecessary function
* lucain's comments.
* fixes
* address high-level comments.
* fix test
* subfolder shenanigans./
* Update src/diffusers/utils/hub_utils.py
Co-authored-by: Lucain <lucainp@gmail.com>
* Apply suggestions from code review
Co-authored-by: Lucain <lucainp@gmail.com>
* remove _huggingface_hub_version as not needed.
* address more feedback.
* add a test for local_files_only=True/
* need hf hub to be at least 0.23.2
* style
* final comment.
* clean up subfolder.
* deal with suffixes in code.
* _add_variant default.
* use weights_name_pattern
* remove add_suffix_keyword
* clean up downloading of sharded ckpts.
* don't return something special when using index.json
* fix more
* don't use bare except
* remove comments and catch the errors better
* fix a couple of things when using is_file()
* empty
---------
Co-authored-by: Lucain <lucainp@gmail.com>
* 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>
