[Gradient Compression] Surface C++ comm hooks to Python API as built-in comm hooks #46959
Conversation
This file contains 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
…in comm hooks 1. Implement the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
requested review from
apaszke,
mingzhe09088,
mrshenli,
pietern,
pritamdamania87,
rohan-varma and
zhaojuanmao
as code owners
This was referenced Oct 27, 2020
facebook-github-bot
added
the
oncall: distributed
💊 CI failures summary and remediationsAs of commit 3f27a7b (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group. This comment has been revised 34 times. |
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 28, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115321037 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 28, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115335197 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 29, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115450614 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 29, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115455858 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
It seems that this version is slightly (1%-2%) slower than the earlier C++ hacks. I suspect that the overheads are due to the usage of future.then() in our C++ comm hook. |
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 29, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115496448 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
Does this mean it is actually on par with the Python Hook or is the python hook slower than this hook? |
pritamdamania87
approved these changes
Oct 30, 2020
Need to run it on Ads 10x to verify. Unfortunately, that training command was somehow broken when I wrote this diff. Will verify it again later. |
Sure. I don't think I can land it before the fix. |
…I as built-in comm hooks" 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Oct 30, 2020
…in comm hooks Pull Request resolved: #46959 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115629230 Differential Revision: [D24471910](https://our.internmc.facebook.com/intern/diff/D24471910/)
|
This pull request has been merged in ee0033a. |
wayi1
pushed a commit
that referenced
this pull request
Nov 3, 2020
…in comm hooks This is almost same as #46959, except that in caffe2/torch/nn/parallel/distributed.py, BuiltinCommHookType is imported conditionally, only when dist.is_available(). Otherwise, this Python enum type defined in caffe2/torch/scrc/distributed/c10d/init.cpp cannot be imported, which is similar to another enum type ReduceOp defined in the same file. See #47153 To review the diff on top of #46959, compare V1 vs Latest. Main Changes in V1 (#46959): 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24700959](https://our.internmc.facebook.com/intern/diff/D24700959/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Nov 3, 2020
…in comm hooks This is almost same as #46959, except that in caffe2/torch/nn/parallel/distributed.py, BuiltinCommHookType is imported conditionally, only when dist.is_available(). Otherwise, this Python enum type defined in caffe2/torch/scrc/distributed/c10d/init.cpp cannot be imported, which is similar to another enum type ReduceOp defined in the same file. See #47153 To review the diff on top of #46959, compare V1 vs Latest. Main Changes in V1 (#46959): 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24700959](https://our.internmc.facebook.com/intern/diff/D24700959/) ghstack-source-id: 115753518 Pull Request resolved: #47270
wayi1
pushed a commit
that referenced
this pull request
Nov 3, 2020
…I as built-in comm hooks" This is almost same as #46959, except that in caffe2/torch/nn/parallel/distributed.py, BuiltinCommHookType is imported conditionally, only when dist.is_available(). Otherwise, this Python enum type defined in caffe2/torch/scrc/distributed/c10d/init.cpp cannot be imported, which is similar to another enum type ReduceOp defined in the same file. See #47153 Main Changes in #46959: 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 Differential Revision: [D24700959](https://our.internmc.facebook.com/intern/diff/D24700959/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Nov 3, 2020
…in comm hooks Pull Request resolved: #47270 This is almost same as #46959, except that in caffe2/torch/nn/parallel/distributed.py, BuiltinCommHookType should be imported conditionally, only when dist.is_available(). Otherwise, this Python enum type defined in caffe2/torch/scrc/distributed/c10d/init.cpp cannot be imported. See #47153 I tried to follow another enum type enum type ReduceOp defined in the same file, but did not work, because the C++ enum class is defined torch/lib/c10d library, but BuiltinCommHookType is defined in torch/csrc/distributed library. These two libraries are compiled in two different ways. To avoid adding typing to distributed package, which can be a new project, I simply removed the arg type of BuiltinCommHookType in this file. To review the diff on top of #46959, compare V1 vs Latest: https://www.internalfb.com/diff/D24700959?src_version_fbid=270445741055617 Main Changes in V1 (#46959): 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115783237 Differential Revision: [D24700959](https://our.internmc.facebook.com/intern/diff/D24700959/)
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 4, 2020
…in comm hooks (#47270) Summary: Pull Request resolved: #47270 This is almost same as #46959, except that in caffe2/torch/nn/parallel/distributed.py, BuiltinCommHookType should be imported conditionally, only when dist.is_available(). Otherwise, this Python enum type defined in caffe2/torch/scrc/distributed/c10d/init.cpp cannot be imported. See #47153 I tried to follow another enum type enum type ReduceOp defined in the same file, but did not work, because the C++ enum class is defined torch/lib/c10d library, but BuiltinCommHookType is defined in torch/csrc/distributed library. These two libraries are compiled in two different ways. To avoid adding typing to distributed package, which can be a new project, I simply removed the arg type of BuiltinCommHookType in this file. To review the diff on top of #46959, compare V1 vs Latest: https://www.internalfb.com/diff/D24700959?src_version_fbid=270445741055617 Main Changes in V1 (#46959): 1. Implemented the Pybind part. 2. In the reducer, once the builtin_comm_hook_type is set, a c++ comm hook instance will be created in Reducer::autograd_hook. 3. Added unit tests for the builit-in comm hooks. Original PR issue: C++ DDP Communication Hook #46348 ghstack-source-id: 115783237 Test Plan: buck test mode/dev-nosan caffe2/test/distributed:c10d -- test_builtin_ddp_comm_hooks_nccl //arvr/projects/eye_tracking/Masquerade:python_test USE_DISTRIBUTED=0 USE_GLOO=0 BUILD_TEST=0 USE_CUDA=1 USE_MKLDNN=0 DEBUG=0 python setup.py install Reviewed By: mrshenli Differential Revision: D24700959 fbshipit-source-id: 69f303a48ae275aa856e6e9b50e12ad8602e1c7a
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:
Original PR issue: C++ DDP Communication Hook #46348
Differential Revision: D24471910