New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
In the CUDA implementation of erfinv, erfinv() should be used for double #25337
Conversation
It is possible to test cover, to make sure that we are not repeating this error again. |
2a2dbb8
to
7a43dfb
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@VitalyFedyunin has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
fef39cf
to
1a88df9
Compare
@VitalyFedyunin Could you re-import please? Now this should be ready. Also, I've added test to ensure the accuracy for double |
@pytorchbot rebase this please |
@VitalyFedyunin Could you re-review this, please? Thanks! |
@pytorchbot rebase this please |
This best preserves accuracy, while erfinvf() should be used for half and float.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nice!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@zou3519 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
…ble (#25337) Summary: This best preserves accuracy, while erfinvf() should be used for half and float. This is also consistent with the implementation before the migration: pytorch/pytorch#24943 Pull Request resolved: pytorch/pytorch#25337 Differential Revision: D17102333 Pulled By: zou3519 fbshipit-source-id: 5178cff534cf5f10d86ab04d4b6c1779ffedf49e
This best preserves accuracy, while erfinvf() should be used for half and float.
This is also consistent with the implementation before the migration: #24943