Test SamplingProbabilityCorrection layer with different dimensions.
#43
+68
−24
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
The `SamplingProbabilityCorrection` already supports different dimensions. This adds testing for it. See keras-team#39
abheesht17
reviewed
Apr 8, 2025
Comment on lines
+32
to
+34
| "testcase_name": "logits_rank_2_probs_rank_1", | ||
| "logits_rank": 2, | ||
| "probs_rank": 1, |
There was a problem hiding this comment.
Will take a closer look, but shouldn't we ensure that ranks for logits and probs are equal?
That is, logits and ranks can have any dimension, but logits and probs should have equal rank
There was a problem hiding this comment.
This is a real use case from what I can see.
If your logits have shape [batch_size, num_candidates], you're allowed to pass a probabilities of shape [num_candidates].
The reason, I assume, is that the probability correction is per candidate regardless of the batch index. To not force you to duplicate the probabilities for each batch item, we allow probabilities to have fewer dimensions.
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Outdated suggestions cannot be applied.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.
The
SamplingProbabilityCorrectionalready supports different dimensions. This adds testing for it.See #39