Fix Tensorflow T5 with int64 input #13479
Merged
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
Rocketknight1
requested review from
patrickvonplaten,
LysandreJik and
sgugger
sgugger
approved these changes
Sep 8, 2021
LysandreJik
approved these changes
Sep 8, 2021
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Outdated suggestions cannot be applied.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.
Our Tensorflow T5 model uses constant values that are initialized with
tf.constant()or just from Python int literals. Tensorflow guesses the dtype in both of these cases asint32, which causes the operations using them to fail ifint64input is supplied.This PR adds checks that cast the constants to the dtype of the inputs, to ensure that the operations succeed for both
int32andint64inputs.