[FSDP()][3/N] Refactor public APIs #87917
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ghstack-poisoned]
awgu
requested review from
H-Huang,
kwen2501,
mrshenli,
pritamdamania87,
rohan-varma and
zhaojuanmao
as code owners
pytorch-bot
bot
added
the
release notes: distributed (fsdp)
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/87917
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e688429: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Oct 27, 2022
This PR is easy. I moved `BackwardPrefetch` to a new file `common_utils.py` and reworded the docs a bit. [ghstack-poisoned]
This PR is easy. I moved `BackwardPrefetch` to a new file `common_utils.py` and reworded the docs a bit. [ghstack-poisoned]
This PR is easy. I moved `BackwardPrefetch` to a new file `common_utils.py` and reworded the docs a bit. [ghstack-poisoned]
awgu
pushed a commit
to awgu/pytorch
that referenced
this pull request
Oct 29, 2022
ghstack-source-id: 6dc3b46 Pull Request resolved: pytorch#87917
This PR is easy. I moved `BackwardPrefetch` to a new file `common_utils.py` and reworded the docs a bit. [ghstack-poisoned]
awgu
changed the title
BackwardPrefetch enum
added 4 commits
October 29, 2022 21:35
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. [ghstack-poisoned]
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. [ghstack-poisoned]
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. [ghstack-poisoned]
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. [ghstack-poisoned]
awgu
pushed a commit
to awgu/pytorch
that referenced
this pull request
Oct 30, 2022
ghstack-source-id: 0ba00bf Pull Request resolved: pytorch#87917
This was referenced Oct 31, 2022
kulinseth
pushed a commit
to kulinseth/pytorch
that referenced
this pull request
Nov 5, 2022
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. Pull Request resolved: pytorch#87917 Approved by: https://github.com/mrshenli
kulinseth
pushed a commit
to kulinseth/pytorch
that referenced
this pull request
Dec 10, 2022
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`. - This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings. - This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs. - To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code. Pull Request resolved: pytorch#87917 Approved by: https://github.com/mrshenli
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack:
_lazy_init()into_fsdp_root_pre_forward()#87941 [FSDP()][26/N] Move_lazy_init()into_fsdp_root_pre_forward()_post_forward_reshard()#87940 [FSDP()][25/N] Add_post_forward_reshard()_lazy_init()#87939 [FSDP()][24/N] Refactor_lazy_init()_reset_lazy_init()#87937 [FSDP] Simplify_reset_lazy_init()_cast_buffers()in_lazy_init()#87936 [FSDP()][22/N] Refactor_cast_buffers()in_lazy_init()_cast_buffers()#87935 [FSDP()][21/N] Refactor_buffer_name_to_orig_dtypecomputationdtypetobuffer_name_to_dtype#87934 [FSDP] Renamedtypetobuffer_name_to_dtypedevicearg from_cast_buffers()#87933 [FSDP] Removedevicearg from_cast_buffers()pre_forward_unshard()#87931 [FSDP()][18/N] Refactorpre_forward_unshard()_fsdp_root_pre_forward()#87930 [FSDP()][17/N] Refactor_fsdp_root_pre_forward()_init_streams()#87928 [FSDP()][15/N] Refactor_init_streams()_FreeEventQueue#87922 [FSDP()][8/N] Refactor limiter's_FreeEventQueueapi.pymeant to hold the public API for FSDP (minusFullyShardedDataParallelitself). This is needed because several of the_<...>_utils.pyfiles rely on the public API, and we cannot import fromtorch.distributed.fsdp.fully_sharded_data_parallelwithout a circular import. Calling the fileapi.pyfollows the convention used byShardedTensor.BackwardPrefetch,ShardingStrategy,MixedPrecision, andCPUOffloaddocstrings.fsdp.rstto have them rendered in public docs.test_public_bindings.py), the aforementioned classes are removed fromfully_sharded_data_parallel.py's__all__. This is technically BC breaking if someone usesfrom torch.distributed.fsdp.fully_sharded_data_parallel import *; however, that does not happen in any of our own external or internal code.