Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

reverse_sequence's inability to accept int32 can break bidirectional_rnn #1198

Closed
alquraishi opened this issue Feb 19, 2016 · 3 comments
Closed
Assignees
Labels

Comments

@alquraishi
Copy link

In the latest releases bidirectional_rnn has been changed to accept int32 tensors for the sequence_length argument, but tf.reverse_sequence only accepts int64 tensors, and this is currently causing an error when an int32 tensor is passed to bidirectional_rnn.

@girving
Copy link
Contributor

girving commented Mar 8, 2016

@ludimagister: Is this problematic to fix?

@girving girving added the bug label Mar 8, 2016
vrv pushed a commit that referenced this issue Mar 9, 2016
@ludimagister
Copy link

Sorry for the delay, this is fixed at head now.

Mike

@girving
Copy link
Contributor

girving commented Mar 9, 2016

@ludimagister: For future use: if you include "Fixes #" in the commit description the bug will be automatically closed on push.

@aselle aselle added type:bug Bug and removed bug labels Feb 9, 2017
fsx950223 pushed a commit to fsx950223/tensorflow that referenced this issue Nov 28, 2023
…upstream-deven-misc-201207-2

Enable XLA custom_call_thunk on the ROCm platform
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

4 participants