[autoparallel] add numerical test for node strategies #1760
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
FrankLeeeee
reviewed
Oct 26, 2022
FrankLeeeee
reviewed
Oct 26, 2022
| @@ -115,6 +115,7 @@ class CommAction: | |||
| comm_spec: CommSpec = None | |||
| comm_type: CommType = None | |||
| arg_index: int = -1 | |||
| key_for_kwarg: any = None | |||
There was a problem hiding this comment.
what is this used for?
There was a problem hiding this comment.
If a node is kwarg of other node, not a positional argument, we could use this to locate it in runtime pass.
FrankLeeeee
reviewed
Oct 26, 2022
FrankLeeeee
reviewed
Oct 26, 2022
tests/test_auto_parallel/test_tensor_shard/test_node_handler/utils.py
Outdated
Show resolved
Hide resolved
FrankLeeeee
reviewed
Oct 26, 2022
tests/test_auto_parallel/test_tensor_shard/test_node_handler/utils.py
Outdated
Show resolved
Hide resolved
FrankLeeeee
reviewed
Oct 26, 2022
FrankLeeeee
approved these changes
Oct 27, 2022
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Outdated suggestions cannot be applied.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.
What does this PR do
This PR provides some apis to do numerical tests for node strategies. The function called numerical_test_for_node_strategy will check the output of forward result, the gradient of the input activation and gradient of parameters if exist with the original result of non-distributed version.
Note: only convolution handler tests updated with the above apis as an example, other handlers unit test will be updated in next PR.