GitLab

This is no longer needed since isort's version is 5.10

Also fix black version to 22.3.0 to fix issue with click
dependency.

Update files that now fail with new version of black {a = 2 ** 4} ->
{a = 2**4}

* update pytest versions

* [test] test related changes

- upgrade to newer pytorch versions
- added function to make test more deterministic on A100 and TF32
- fixed some tests so that they are correctly skipped on a single GPU system

* more fixes

* formatting overly long lines

* format

* better test without trigger a warning

* fix an optim state bug with newer pytorch

- adam optimizer seems to return "step" as a singleton tensor now in the
nightly build
- this fixes it assumeing non-tensor value can still be loaded back by
the optimizer

* improve oss.py

- use min_loss for regression checking is a bit more reliable
- also increased the num epochs from 10 to 12

* small oss.py fix

* Update fairscale/nn/data_parallel/fully_sharded_data_parallel.py
Co-authored-by: Min Xu <min.xu.public@gmail.com>

* skipping one more test

* formatting

* minor fix and copyright header

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

* [skip ci] first commit

* [skip ci] gradient scaler example

* [skip ci] adding feed forward toy example

* [skip ci] adding types

* [skip ci] adding backward hook

* [skip ci] update

* [skip ci] working feed forward example

* [skip ci] working feed forward example

* [skip ci] use named_modules instead of named_children

* [skip ci] adding new file

* [skip ci] clean up

* [skip ci] implement unscale function

* [skip ci] implement unscale function

* [skip ci] removing old file

* [skip ci] removing some more old files

* [skip ci] making unscale function generic

* [skip ci] adding test for vision model

* [skip ci] adding identity layer

* [skip ci] cleanup files

* [skip ci] refactoring

* [skip ci] more refactoring

* [skip ci] added functionality to update scale

* [skip ci] data loader clean up

* [skip ci] implemented inf checks and update scale functions

* [skip ci]code clean up. added...

* adding pre-commit files

* applying pre-commit to all files

* adding no-strict-optional argument to mypy in circle ci config

* fix typo

* updating python versions

* [skip ci] remove extra args

* adding python 3.9

* [skip ci] set pre-commit version in requirements-dev.txt

* set CACHE_VERSION

* move linters from circleci to github actions

* update python version

* update python version in benchmarks_2

* moving to python 3.9.7

[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>

* rename and move optim/utils.py

* attach the new file

* making APIs more private
* linting

* param buckets
* unifying the buckets

* enabling disabled tests

* extending the current state_dict interface, make it possible to do everything in a single call, and to checkpoint on all ranks

* Adding a hard sync barrier before the broadcast, mostly useful for Gloo actually, NCCL is synced behind the scene
* adding a proper unit test
* adding a unit test for https://github.com/facebookresearch/fairscale/pull/510

* change empty shard handling for OSS, do not rely on asserts
* code review

- cover them in terms of code path only
- numerically, AdaScale is different on SDP/FSDP than DDP, mainly
  due to partial view of the gradients.
- this doesn't mean it is definitely not useful but it is yet to
  be validated.
- not going to spend too much time until we have a real use case.

Recent work by [Microsoft](https://arxiv.org/abs/1910.02054) and [Google](https://arxiv.org/abs/2004.13336

) has shown that data parallel training can be made significantly more efficient by sharding the model parameters and optimizer state across data parallel workers. These ideas are encapsulated in the new **`FullyShardedDataParallel` (FSDP)** wrapper, which is a drop-in replacement for PyTorch's `DistributedDataParallel` (DDP) wrapper.

Compared to PyTorch DDP:
* FSDP shards parameters (FP16 + FP32) and optimizer state across data parallel GPUs
* FSDP with `reshard_after_forward=False` has the same communication cost as PyTorch DDP and is similar to ZeRO-2
* FSDP with `reshard_after_forward=True` increases total communication by 50% and is similar to ZeRO-3:
    * all-gather parameters at start of forward pass and start of backward pass
    * reduce-scatter grads at end of backward pass
Co-authored-by: Min Xu <24926999+min-xu-ai@users.noreply.github.com>
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>

* adding an assert + corresponding unit test
* updated changelog
* adjusting the adascale tests

* WIP, needs to be fixed !

* should be a fix, many thanks Weiyi Zheng

* slightly better unit test, sorting the states on the way out

* reproducing the issue from Weiyi in a unit test, and finally properly fixing

* fixing unit test on pytorch1.5 - original loss diff 26.404895782470703 - 26.404342651367188

* Better unit testing
* Make it possible to refresh the DDP assumptions when the model has changed. Make it optional so that you save some time
* Enabling accumulation tests

fix a broken earlier commit, only worked for the first step

* precise skip, only if agent has only cpu

* [feat] Add AdaScaleWrapper

- This enables a different API for wrapping an optimizer with AdaScale.
- This also enables AdaScale to be wrapped by OSS.
- However, OSS wrapping AdaScale results in different optimization,
  which future research will be needed to study its effects.

testing: add unit tests.

* addressed comment: typo

* adding a test to prove the inter operability with upstream pytorch
* updating the changelog
* eager state pruning
* pytorch 1.5 compat

* [test]: test with py39 + torch 1.8 nightly

* version fix

* more fix

* fix version function for nightly version

* fix torch_pg build

* invalidate cache

* separate benchmark requirements

* comment

* fixed mypy

* fixed a test

* [test]: test adascale with oss

* minor fix

* add a small comment

* refactor: moved find_tensor_by_shape

* refactor: move test golden data into its own module

* refactor: simplied the train function

* refactor: added comments as suggested

* tentatively fixing the cpu version of circleci jobs, now pipe tests are the last ones standing
* fixing oss backcompat, trying to fix rpc in old pytorch also
* fixing the file based init in torch 1.5

* adding a parity unit test
* code review, better testing, use torch defaults and check for the loss, log world size