Split nn.MultiHeadAttention into Module + functional #20415
Conversation
cpuhrsch
changed the title
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@jamarshon 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.
@jamarshon 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.
@jamarshon 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.
@jamarshon has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
cpuhrsch
left a comment
cpuhrsch
left a comment
There was a problem hiding this comment.
Once all tests pass this should be good to. cc @zhangguanheng66
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@jamarshon has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
|
Current test failures are due to temporary CI outages and restricted to AMD and Windows only. |
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@jamarshon has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
jamarshon
changed the title
|
@jamarshon merged this pull request in 6e82b1c. |
6 participants
Moving functions from torch/nn/modules/activation.py to torch/nn/functional.py. For functions not implemented (_get_input_buffer and _set_input_buffer), a TODO is added.