Enable roberta embedding #786
Open
Conversation
This file contains 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
yeonsily
requested review from
kzawora-intel,
madamczykhabana,
michalkuligowski,
mgawarkiewicz,
vivekgoe and
afierka-intel
as code owners
yeonsily
force-pushed
the
dev/enable_roberta_embedding
branch
from
594a1d3 to
c660333
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
We set position_ids and input_ids as [batch_size, bucket_size] on hpu so need to modify the current roberta embedding forward function.
e.g. position_id on gpu
[0,1,2,3,4,5,6,7]
but position_id on hpu
[0,1,2,3,4,5,6,7,0,0,0,.....0,0] which size is 128 with padding in this case.
One thing I noticed is torch.equal() on hpu is not working properly and I have to run it on cpu.
That code is nothing but checking pre-condition but we will need to investigate it.
This PR has a dependency on #758