fix: grad for indexed ops with axes #1360
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
polvalente
commented
Nov 9, 2023
nx/lib/nx/binary_backend.ex
Outdated
Comment on lines
2107
to
2112
| inverse_permutation = | ||
| if diff > 0 do | ||
| Enum.to_list(0..(diff - 1)) ++ Enum.map(inverse_permutation, &(&1 + diff)) | ||
| else | ||
| inverse_permutation | ||
| end |
There was a problem hiding this comment.
I'm not totally sure if this is the correct fix. Basically the test for grad(gather) was failing and this fixed it
There was a problem hiding this comment.
Will we ever hit the else branch?
There was a problem hiding this comment.
It seems we can. Keeping only the do branch works if we change the range to 0..(diff-1)//1 because it seems that for some cases the diff is 0. I think those are cases where the index tensor as many axes as the input tensor.
The Enum.filter(... in Nx.axes(out)) takes care of the case where the index tensor has fewer axes, and the do block takes care of when there are more axes than in the input.
josevalim
approved these changes
Nov 11, 2023
|
💚 💙 💜 💛 ❤️ |
josevalim
pushed a commit
that referenced
this pull request
Nov 11, 2023
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Outdated suggestions cannot be applied.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.
No description provided.