Fix copy_transpose_valid check #20759
Conversation
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@colesbury has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@colesbury has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
|
@pytorchbot rebase this please |
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@colesbury has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
|
:) |
| self.assertEqual(y[:, 40], range(4000, 4100)) | ||
|
|
||
| y = torch.empty(100, 100, dtype=torch.double) | ||
| y.copy_(x) |
There was a problem hiding this comment.
@colesbury This tests copy from a transposed float array (x) to a contiguous float/double array (y). Though in my example it's double->float direction that breaks. Not sure it changes much though
There was a problem hiding this comment.
The test captures the aspect that it's copying between different data types, which is all that matters here.
|
@colesbury merged this pull request in 7a0c6d5. |
Summary: Fixes #20755 (Was broken in #20685) cc vadimkantorov Pull Request resolved: pytorch/pytorch#20759 Differential Revision: D15433712 Pulled By: colesbury fbshipit-source-id: 29f612f7d4d7b73158d6f5dc1e46fd2f8fb09a2f
6 participants
Fixes #20755
(Was broken in #20685)
cc @vadimkantorov