GitLab

* update python version for cpu tess

* run CPU tests with updated PyTorch version

* update nightly and test PyTorch versions

* skip failing multiprocess pipe test

* always skip test

* always skip test

* always skip test

* lint error

* skip unsupported versions

* improve skip message

* lint errors

* add log for new memory tracker features

* add log for new memory tracker features

* [feat] layer memory tracking

* [feat] layer memory tracking (add tests in CI)

* [feat] layer memory tracking: doc typos

* [feat] layer memory tracking: mypy fixes

* [feat] layer memory tracking: fixes for FSDP all gather tracking on pytorch 1.9 and above

* [feat] layer memory tracking: lint

* [feat] layer memory tracking: mypy
Co-authored-by: QuentinDuval <QuentinDuval@users.noreply.github.com>

* fix

* remove dup file

* update master branch to main

* added FAQ about updating the branch from master to main

* fixed some false positive correction

* added what is new section

* fixed the quoted code area

* added release what is new section

* added a step in release.md

* fixed a word

* [chore]0.4.1 release

* put more details in one change log

* add toggler to disable the using the nccl base collectives

* added todo to remove the toggle when the issue is resolved.

* add changelog for previous commit

* add changelog for previous commit

* add changelog for previous commit

* fix a merge induced error
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* [fix] FSDP intra-backwards gradient accumulation.

Ensure gradient reduction accumulates into the unsharded gradient tensor
within a backwards pass. This matters when an FSDP module is called
multiple times within a forward pass, and reduction is _not_ deferred
using activation checkpoint forward counters, bucketing or some other
mechanism.

Closes #780

* [refactor] Remove forward counters. Comments.

Removed forward counters from the activation checkpointing utility, now
that FSDP does not require them for correct operation. Add more detailed
comment about memory usage behaviour with gradient reduction.

* [refactor] Delete deprecated forward counter usage.

* [refactor] Add state assertion as end of pre-backward hook.

Before this commit, output tensors of checkpointed modules always
require grad, even if they shouldn't. This commit makes it so that
the outputs of checkpointed modules only require grad if either
the input requires grad or if the parameters require grad.

To achieve this, this commit also adds a new _unflattened_param_views
attribute to modules being flattened. This allows the checkpointing
to still access the parameters and check if gradients need to be
computed.
Co-authored-by: Alex Xiao <axiao@fb.com>

Co-authored-by: Min Xu <min.xu.public@gmail.com>

* [test] Added disable_checkpointing unit test

* [test] Added disable_checkpointing unit test (Clean-up)

* [test] Added disable_checkpointing unit test (Clean-up)

See https://github.com/facebookresearch/fairscale/issues/771

[cleanup] CI test updates; mypy cleanup; partial broadcast_object cleanup; pre-commit documentation (#744)

* changelog; mypy; oss cleanup

* more broadcast_object cleanup in FSDP

* one more mypy fix

* retire pytorch 1.6 from circleci, add new lightly, add 1.8 LTS and 1.9 stable release

* update torch version for LTS

* minor fixes

* update cache key

* trying newer gpu VMs

* bump the cache

* update to gpu.medium, which should be 2 GPUs

* update nightly version

* add pre-commit instruction

* fixed CHANGELOG after merging

* updated to newer nightly

* retained the older broadcast function for older GPUs for oss.py

* fixed a bug

* added a comment

* fixing a test for pytorch 1.10

* testing a fix

* Update fairscale/optim/oss.py

* Update CONTRIBUTING.md
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* [bug] [FSDP] making sure we use full params for multiple backwards within an iteration

* changelog
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* minor: changelog and pre-commit

* addressed comment

* update the release doc
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* add additional assert for checking if the requires_grad field is set.

* fix lint errors

* add unit tests and address comments

[FSDP][feature] Support returning the original parameter names after a model has been wrapped with FSDP (#755)

* checkpoint work

* fix lint issues

* remove debug statement

* remove print

* fix lint errors

* fix lint errors

* fix lint errors

* add comments and fix lint errors

* modified comments and tests

Pre-commit hook fails when run on all files for three reasons:
(see trace below)

1. Trailing whitespace on multiple files
2. mypy fails to load numpy and then subsequently fails to load
LazyModule from pipe.py
3. isort sees issues with known_third_party packages

```
> pre-commit run --all-files

Trim Trailing Whitespace.................................................Failed
- hook id: trailing-whitespace
- exit code: 1
- files were modified by this hook

Fixing docs/source/conf.py
Fixing fairscale/experimental/nn/auto_shard.py
Fixing docs/source/deep_dive/activation_checkpointing.rst
Fixing docs/source/tutorials/pipe.rst
Fixing docs/source/installation_instructions.rst
Fixing docs/source/deep_dive/pipeline_parallelism.rst
Fixing docs/source/tutorials/activation_checkpointing.rst
Fixing docs/source/tutorials/offload_model.rst
Fixing docs/source/deep_dive/oss_sdp_fsdp.rst
Fixing docs/source/what_is_fairscale.rst
Fixing CHANGELOG.md
Fixing fairscale/experimental/nn/offload.py
Fixing docs/source/index.rst
Fixing docs/source/deep_dive/adascale.rst
Fixing README.md
Fixing docs/source/tutorials/oss.rst
Fixing docs/source/deep_dive/offload.rst

Check python ast.........................................................Passed
Check for merge conflicts................................................Passed
Don't commit to branch...................................................Passed
Check for added large files..............................................Passed
Fix End of Files.........................................................Failed
- hook id: end-of-file-fixer
- exit code: 1
- files were modified by this hook

Fixing requirements.txt
Fixing docs/source/getting_started.rst
Fixing docs/source/installation_instructions.rst
Fixing codecov.yml
Fixing docs/source/deep_dive/adascale.rst
Fixing docs/source/tutorials/oss.rst
Fixing docs/source/deep_dive/offload.rst

black....................................................................Passed
flake8...................................................................Passed
seed isort known_third_party.............................................Failed
- hook id: seed-isort-config
- exit code: 1
- files were modified by this hook
isort....................................................................Passed
mypy.....................................................................Failed
- hook id: mypy
- exit code: 2

setup.cfg:45: error: Error importing plugin 'numpy.typing.mypy_plugin': No module named 'numpy'
Found 1 error in 1 file (checked 197 source files)
```

`wrap` from `auto_wrap` is used in the docstring example which is missing from the imports.

Co-authored-by: Min Xu <min.xu.public@gmail.com>

* Add test (broken) for gradient accumulation without no_sync context manager

* changelog

* no_sync to grad_acc renaming for tests

* clean up tmp files

* support grad acc without no_sync

* minor

* update changelog

* Update fairscale/nn/data_parallel/fully_sharded_data_parallel.py

Better assertion from Sam.
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>

* lint
Co-authored-by: Min Xu <min.xu.public@gmail.com>
Co-authored-by: Min Xu <24926999+min-xu-ai@users.noreply.github.com>
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>

Move final backward callback to pre-backward hook of root FSDP instance

Summary:

Move final backward callback to pre-backward hook of root FSDP instance,
so that it is always attached to the outer most backward call and fired
after all backward calls are completed.

Also added flags to check final backward callback is fired when final
backward callback is required.

If root FSDP is checkpointed and called multiple times in forward,
check pointer counter is used to make sure final backward callback is queued inside last inner backward
call as well.

Test Plan: unit tests

Reviewers:

Subscribers:

Tasks:

Tags:

* reformat

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* nits and unit tests

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* address some comments

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* replace m with self
Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* reformat

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* nits

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* remove the fired flag

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* assert state on root only

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* comments

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* comments

Summary:

Test Plan:

Reviewers:

Subscribers:

Tasks:

Tags:

* [chore] 0.3.9 release

* update changelog

* address comments
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* [feat] FSDP: supporting multiple flatten parameter groups

- step 3: make FSDP use FlattenParamModule unconditionally

* fixing the auto_wrap tests

* minor

* rewrite local_metadata_dict

- updated FPW so that custom flat param name is also supported

* bug fix

* mypy

* rewrote consolidate_shard_weights

- test_consolidate passes

* comments

* fixing pickling

* Fix shared params and MoE logic (#749)

* add strict kwarg to support fairseq:gshard MoE saving logic

* Test fairseq style shard

* style

* formatting and address comments

* added changelog

* fixing a test after padding renaming
Co-authored-by: Min Xu <min.xu.public@gmail.com>
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>

* Update fully_sharded_data_parallel.py

update fully_sharded_data_parallel to use _allgather_base

* Update reduce_scatter_bucketer.py

Use reduce_scatter_base

* Update fully_sharded_data_parallel.py

nonblocking gradient cpu copy, and nonblocking param rebulds

* Update reduce_scatter_bucketer.py

lints

* Update fully_sharded_data_parallel.py

* Update reduce_scatter_bucketer.py

* Update reduce_scatter_bucketer.py

* lints

* linter, test fix

* linter

* LINTERgit add fairscale/utils/reduce_scatter_bucketer.pygit add fairscale/utils/reduce_scatter_bucketer.py

* LINTERgit add tests/nn/data_parallel/test_fsdp_overlap.pygit add tests/nn/data_parallel/test_fsdp_overlap.py

* Update test_fsdp_overlap.py

* Update fairscale/utils/reduce_scatter_bucketer.py
Co-authored-by: Min Xu <24926999+min-xu-ai@users.noreply.github.com>

* Update fairscale/nn/data_parallel/fully_sharded_data_parallel.py
Co-authored-by: Min Xu <24926999+min-xu-ai@users.noreply.github.com>

* Update reduce_scatter_bucketer.py

* isort
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-185.ec2.internal>
Co-authored-by: Min Xu <24926999+min-xu-ai@users.noreply.github.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-77-164.ec2.internal>

misspelled name

See https://github.com/pytorch/pytorch/pull/59671/

Signed-off-by: Edward Z. Yang <ezyang@fb.com>