GitLab

* 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>

* disable peft input autocast

* use new peft method name; only disable peft input autocast if submodule layerwise casting active

* add test; reference PeftInputAutocastDisableHook in peft docs

* add load_lora_weights test

* casted -> cast

* Update tests/lora/utils.py

* start pyramid attention broadcast

* add coauthor
Co-Authored-By: Xuanlei Zhao <43881818+oahzxl@users.noreply.github.com>

* update

* make style

* update

* make style

* add docs

* add tests

* update

* Update docs/source/en/api/pipelines/cogvideox.md
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/api/pipelines/cogvideox.md
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Pyramid Attention Broadcast rewrite + introduce hooks (#9826)

* rewrite implementation with hooks

* make style

* update

* merge pyramid-attention-rewrite-2

* make style

* remove changes from latte transformer

* revert docs changes

* better debug message

* add todos for future

* update tests

* make style

* cleanup

* fix

* improve log message; fix latte test

* refactor

* update

* update

* update

* revert changes to tests

* update docs

* update tests

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

* update

* fix flux test

* reorder

* refactor

* make fix-copies

* update docs

* fixes

* more fixes

* make style

* update tests

* update code example

* make fix-copies

* refactor based on reviews

* use maybe_free_model_hooks

* CacheMixin

* make style

* update

* add current_timestep property; update docs

* make fix-copies

* update

* improve tests

* try circular import fix

* apply suggestions from review

* address review comments

* Apply suggestions from code review

* refactor hook implementation

* add test suite for hooks

* PAB Refactor (#10667)

* update

* update

* update

---------
Co-authored-by: DN6 <dhruv.nair@gmail.com>

* update

* fix remove hook behaviour

---------
Co-authored-by: Xuanlei Zhao <43881818+oahzxl@users.noreply.github.com>
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: DN6 <dhruv.nair@gmail.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>