Skip to content

Conversation

Craigacp
Copy link
Collaborator

This improves determinism and makes the gradients be computed correctly. As it's a control dependency on all the gradients to the consumers of the gradients it's not clear why this is necessary, but it's working.

Fix developed by @nfeybesse.

…rs. This improves determinism and makes the gradients be computed correctly for unclear reasons.
@Craigacp Craigacp changed the base branch from bazelcism to staging February 16, 2024 20:15
@karllessard karllessard merged commit 2bcb8ac into tensorflow:staging Feb 17, 2024
karllessard pushed a commit that referenced this pull request Feb 17, 2024
This improves determinism and makes the gradients be computed correctly for unclear reasons. (#520)

Co-authored-by: Nicolas Feybesse (nfeybesse@users.noreply.github.com)
karllessard pushed a commit that referenced this pull request Feb 23, 2024
This improves determinism and makes the gradients be computed correctly for unclear reasons. (#520)

Co-authored-by: Nicolas Feybesse (nfeybesse@users.noreply.github.com)
karllessard pushed a commit that referenced this pull request Feb 24, 2024
This improves determinism and makes the gradients be computed correctly for unclear reasons. (#520)

Co-authored-by: Nicolas Feybesse (nfeybesse@users.noreply.github.com)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants