[PyTorch] fix(Batch Normalization): memory leaky #1342
Merged
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
|
I just modified the code in pytorch version. |
|
Job d2l-en/PR-1342/2 is complete. |
|
Job d2l-en/PR-1342/3 is complete. |
|
Job d2l-en/PR-1342/4 is complete. |
AnirudhDagar
approved these changes
Aug 15, 2020
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Outdated suggestions cannot be applied.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.
The memory would not be freed and accumulate every epochs to explode finally.
X is associated with moving_mean and moving_var via gradient computation.
As the result, X will not be freed due to the resident of self.moving_mean and self.moving_var in memory.
Thus the returning values here should be modified to moving_mean.data and moving_var.data to remove the gradient property, which will not impact our model.