Support full [b]float16 in embedding_lookup_sparse#49604
Merged
copybara-service[bot] merged 2 commits intotensorflow:masterfrom Jun 8, 2021
Merged
Support full [b]float16 in embedding_lookup_sparse#49604copybara-service[bot] merged 2 commits intotensorflow:masterfrom
copybara-service[bot] merged 2 commits intotensorflow:masterfrom
Conversation
- This removes a forced cast to float32 and instead outputs the same type as the input. The inner computations are still done in float32 to avoid numerical issues. - This improves performance and makes the op consistent with all other ops that output the same type as the input.
google-ml-butler
bot
added
kokoro:force-run
rthadur
added
ready to pull
gbaned
approved these changes
May 28, 2021
google-ml-butler
bot
added
kokoro:force-run
Contributor
Author
|
@penpornk I noticed that |
pkanwar23
added
ready to pull
Contributor
Author
|
Gentle ping |
Member
|
@benbarsdell Sorry for the late reply! Please submit another PR. |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
embedding_lookup_sparseand instead outputs the same type as the input. The inner computations are still done in float32 to avoid numerical issues.cc @nluehr @reedwm