Benchmarks - Keep BatchNorm as fp32 for pytorch cnn models cast to fp16 #322
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
cp5555
changed the title
cp5555
added
benchmarks
abuccts
reviewed
Mar 4, 2022
|
/azp run |
|
Azure Pipelines successfully started running 3 pipeline(s). |
Codecov Report
@@ Coverage Diff @@
## main #322 +/- ##
==========================================
+ Coverage 88.64% 88.66% +0.01%
==========================================
Files 76 76
Lines 4500 4507 +7
==========================================
+ Hits 3989 3996 +7
Misses 511 511
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
cp5555
approved these changes
Mar 4, 2022
guoshzhao
approved these changes
Mar 5, 2022
|
/azp run |
|
Azure Pipelines successfully started running 3 pipeline(s). |
Closed
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Outdated suggestions cannot be applied.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.
Description
The BatchNorm operator is not numerically stable in fp16. PyTorch documentation recommends to keep the BN op in fp32 for fp16 AMP models. Refer to https://pytorch.org/docs/stable/amp.html#ops-that-can-autocast-to-float32. Preserving BN in fp32 for superbench more accurately reflects real workloads.