Skip to content
This repository has been archived by the owner on Oct 17, 2021. It is now read-only.

Use backward layer for computing yRev even when an initial state is given. #318

Merged
merged 3 commits into from
Sep 13, 2018

Conversation

adarob
Copy link
Member

@adarob adarob commented Sep 9, 2018

This change is Reviewable

@bileschi bileschi requested a review from caisq September 11, 2018 13:44
@bileschi
Copy link
Contributor

Thanks @adarob , it would have been nice if this bug had been exposed in a test. Would it be possible / difficult to add a test that would be broken but for your fix?

@caisq
Copy link
Contributor

caisq commented Sep 11, 2018

@adarob, +1 to what @bileschi said. If you are willing to add the test that would be great. We'll also be happy to merge this PR first and add a test ourselves. Just let us know. Thanks.

@caisq
Copy link
Contributor

caisq commented Sep 13, 2018

@bileschi Given we are planning a new minor release soon, I'll merge this PR first and add tests for it later. How does that sound?

@bileschi
Copy link
Contributor

@caisq works for me.

@caisq caisq merged commit 3fa229f into tensorflow:master Sep 13, 2018
@caisq
Copy link
Contributor

caisq commented Sep 13, 2018

@bileschi Thanks.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
3 participants