[autoparallel] Patch meta information of torch.nn.LayerNorm#2647
Merged
YuliangLiu0306 merged 58 commits intohpcaitech:mainfrom Feb 10, 2023
Merged
[autoparallel] Patch meta information of torch.nn.LayerNorm#2647YuliangLiu0306 merged 58 commits intohpcaitech:mainfrom
torch.nn.LayerNorm#2647YuliangLiu0306 merged 58 commits intohpcaitech:mainfrom
Conversation
Merge ColossalAI
Daily merge
Contributor
|
The code coverage for the changed files is 27%. Click me to view the complete report |
Contributor
|
The code coverage for the changed files is 29%. Click me to view the complete report |
YuliangLiu0306
approved these changes
Feb 10, 2023
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
📌 Checklist before creating the PR
[doc/gemini/tensor/...]: A concise description🚨 Issue number
Resolved #2632
📝 What does this PR do?
In this PR, I patch the meta information of
torch.nn.LayerNorm, and I add a function inutils.pyunder thetests/test_auto_parallel/test_tensor_shard/test_metainfodirectory for meta information test.TODO: we could refactor tests under
test_metainfoin the future, as we have meta information patches for communication nodes, there is no need to test meta information patches with clumsy SPMD solver codes. I also move all the tests concerning meta information patches of normalization intotest_norm_metainfo.py.The test will be skipped if
PyTorchversion is lower than 1.12.0, here is a snapshot of test results onPyTorch 1.12.0💥 Checklist before requesting a review
⭐️ Do you enjoy contributing to Colossal-AI?
Tell us more if you don't enjoy contributing to Colossal-AI.