fixed a newly introduced regression in softmax #10066
Closed
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
ktarplee
requested review from
apaszke,
colesbury,
ezyang,
gchanan,
soumith and
zdevito
as code owners
Contributor
|
this is bad, sorry about the regression. Would you be up for adding a test in |
soumith
approved these changes
Jul 31, 2018
Author
|
I would be happy to add to test_nn.py however it looks like somewhat of a pain to build and test pytorch from source. |
Contributor
|
@ktarplee no worries, we'll add it and push this PR through. |
Collaborator
|
Thanks for the fix. Can you also instead submit this PR to the |
Contributor
|
@ssnl i've changed it to be on master. |
soumith
approved these changes
Jul 31, 2018
soumith
approved these changes
Jul 31, 2018
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
soumith is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
goodlux
pushed a commit
to goodlux/pytorch
that referenced
this pull request
Aug 15, 2018
Summary: There is a regression in softmin in 0.4.1 that was not present in 0.4.0. The behavior of softmin(x) should match softmax(-x) however instead it is implemented (in v0.4.1) as -softmax(x). These are not the same. The fix is trivial because the bug is due to operator precedence. This is a major regression that broke my training. I'm not sure how a unit test did not catch this. ``` x = torch.tensor([1, 2, 3.5, 4]) print(F.softmin(x, dim=0)) # this has the wrong output in 0.4.1 but correct in 0.4.0 print(F.softmax(-x, dim=0)) # this is what softmax should be print(F.softmax(x, dim=0)) print(-F.softmax(x, dim=0)) # this is how softmax is implemented incorrectly ``` In 0.4.1 this produces tensor([-0.0278, -0.0755, -0.3385, -0.5581]) tensor([0.6668, 0.2453, 0.0547, 0.0332]) tensor([0.0278, 0.0755, 0.3385, 0.5581]) tensor([-0.0278, -0.0755, -0.3385, -0.5581]) In 0.4.0 this produces the correct values tensor([ 0.6668, 0.2453, 0.0547, 0.0332]) tensor([ 0.6668, 0.2453, 0.0547, 0.0332]) tensor([ 0.0278, 0.0755, 0.3385, 0.5581]) tensor([-0.0278, -0.0755, -0.3385, -0.5581]) Pull Request resolved: pytorch#10066 Differential Revision: D9106995 Pulled By: soumith fbshipit-source-id: 7332503c6077e8461ad6cd72422c749cf6ca595b
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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.
There is a regression in softmin in 0.4.1 that was not present in 0.4.0. The behavior of softmin(x) should match softmax(-x) however instead it is implemented (in v0.4.1) as -softmax(x). These are not the same. The fix is trivial because the bug is due to operator precedence.
This is a major regression that broke my training. I'm not sure how a unit test did not catch this.
In 0.4.1 this produces
tensor([-0.0278, -0.0755, -0.3385, -0.5581])
tensor([0.6668, 0.2453, 0.0547, 0.0332])
tensor([0.0278, 0.0755, 0.3385, 0.5581])
tensor([-0.0278, -0.0755, -0.3385, -0.5581])
In 0.4.0 this produces the correct values
tensor([ 0.6668, 0.2453, 0.0547, 0.0332])
tensor([ 0.6668, 0.2453, 0.0547, 0.0332])
tensor([ 0.0278, 0.0755, 0.3385, 0.5581])
tensor([-0.0278, -0.0755, -0.3385, -0.5581])