C++ API TransformerEncoderLayer #42633
Conversation
[ghstack-poisoned]
glaringlee
requested review from
ebetica,
goldsborough and
yf225
as code owners
glaringlee
requested review from
zhangguanheng66
and removed request for
ebetica and
goldsborough
💊 CI failures summary and remediationsAs of commit 0f1e68f (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 1 failure confirmed as flaky and can be ignored:
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group. This comment has been revised 12 times. |
[ghstack-poisoned]
|
@zhangguanheng66 Can you take a look at the logic in this C++ impl? thx |
[ghstack-poisoned]
|
|
||
| // gelu test case 2 | ||
| encoder_input = torch::tensor({ | ||
| {{0.7462, 0.6653, 0.5679, 0.4891}, {0.5387, 0.1655, 0.3565, 0.0471}}, |
There was a problem hiding this comment.
A quick question. Are those deterministic tests consistent with those in python version?
There was a problem hiding this comment.
@zhangguanheng66 yes, I copied them from python test directly, same number, same precision.
[ghstack-poisoned]
|
@glaringlee merged this pull request in 98de150. |
|
TransformerEncoderLayer options do not match current Pytorch options, particularly batch_first and norm_first. |
Stack from ghstack:
Differential Revision: D22994332