Accumulate MSELoss reduce=True into accreal instead of real #9287
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
zou3519
requested review from
apaszke,
colesbury,
ezyang,
gchanan,
soumith and
zdevito
as code owners
soumith
approved these changes
Jul 9, 2018
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@zou3519 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
zdevito
pushed a commit
to zdevito/ATen
that referenced
this pull request
Jul 11, 2018
Summary: THNN was accumulating the result of reduction loss functions into real instead of accreal. This was causing precision issues with MSELoss. This patch only fixes MSELoss. Some of the other losses exhibit bad precision as well (because they accumulate into real instead of accreal) and require more investigation. I will open an issue for those (#9286) Fixes #8710 cc li-roy SsnL Pull Request resolved: pytorch/pytorch#9287 Reviewed By: SsnL Differential Revision: D8775708 Pulled By: zou3519 fbshipit-source-id: d1a1f159deee0cb90fd8e81e63b246115eea8e9e
zdevito
pushed a commit
to zdevito/ATen
that referenced
this pull request
Jul 13, 2018
Summary: THNN was accumulating the result of reduction loss functions into real instead of accreal. This was causing precision issues with MSELoss. This patch only fixes MSELoss. Some of the other losses exhibit bad precision as well (because they accumulate into real instead of accreal) and require more investigation. I will open an issue for those (#9286) Fixes #8710 cc li-roy SsnL Pull Request resolved: pytorch/pytorch#9287 Reviewed By: SsnL Differential Revision: D8775708 Pulled By: zou3519 fbshipit-source-id: d1a1f159deee0cb90fd8e81e63b246115eea8e9e
goodlux
pushed a commit
to goodlux/pytorch
that referenced
this pull request
Aug 15, 2018
…9287) Summary: THNN was accumulating the result of reduction loss functions into real instead of accreal. This was causing precision issues with MSELoss. This patch only fixes MSELoss. Some of the other losses exhibit bad precision as well (because they accumulate into real instead of accreal) and require more investigation. I will open an issue for those (pytorch#9286) Fixes pytorch#8710 cc li-roy SsnL Pull Request resolved: pytorch#9287 Reviewed By: SsnL Differential Revision: D8775708 Pulled By: zou3519 fbshipit-source-id: d1a1f159deee0cb90fd8e81e63b246115eea8e9e
gchanan
added a commit
that referenced
this pull request
Aug 28, 2020
These were (apparently) accidentally turned off two years ago in #9287. [ghstack-poisoned]
gchanan
added a commit
that referenced
this pull request
Aug 30, 2020
These were (apparently) accidentally turned off two years ago in #9287. Differential Revision: [D23408987](https://our.internmc.facebook.com/intern/diff/D23408987)
gchanan
added a commit
that referenced
this pull request
Aug 31, 2020
These were (apparently) accidentally turned off two years ago in #9287. Differential Revision: [D23408987](https://our.internmc.facebook.com/intern/diff/D23408987) [ghstack-poisoned]
gchanan
added a commit
that referenced
this pull request
Sep 2, 2020
This looks to have been a mistake from #9287. [ghstack-poisoned]
gchanan
added a commit
that referenced
this pull request
Sep 2, 2020
This looks to have been a mistake from #9287. [ghstack-poisoned]
gchanan
added a commit
that referenced
this pull request
Sep 2, 2020
This looks to have been a mistake from #9287. Differential Revision: [D23476274](https://our.internmc.facebook.com/intern/diff/D23476274) [ghstack-poisoned]
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
THNN was accumulating the result of reduction loss functions
into real instead of accreal. This was causing precision issues with
MSELoss.
This patch only fixes MSELoss. Some of the other losses exhibit bad precision as well (because they accumulate into real instead of accreal) and require more investigation. I will open an issue for those (#9286)
Fixes #8710
cc @li-roy @ssnl