Add min_examples_per_batch to FeatureAblation/FeaturePermutation #1533
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
|
This pull request was exported from Phabricator. Differential Revision: D71980024 |
sarahtranfb
added a commit
to sarahtranfb/captum
that referenced
this pull request
Mar 28, 2025
…orch#1533) Summary: For sparse features, the 0th dim is not always the batch size. Currently for FeatureAblation, we skip the input tensor if there are no elements in it. For FeaturePermutation, we throw if we cannot permute across the "batch" (i.e. the tensor has only one element). `min_examples_per_batch` parameterizes the above behavior. For FeaturePermutation, the default is set to 2, meaning we no longer throw by default, but return an attribution score of 0 Reviewed By: jjuncho Differential Revision: D71980024
sarahtranfb
force-pushed
the
export-D71980024
branch
from
1ff3c49 to
5959648
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71980024 |
…orch#1533) Summary: For sparse features, the 0th dim is not always the batch size. Currently for FeatureAblation, we skip the input tensor if there are no elements in it. For FeaturePermutation, we throw if we cannot permute across the "batch" (i.e. the tensor has only one element). `min_examples_per_batch` parameterizes the above behavior. For FeaturePermutation, the default is set to 2, meaning we no longer throw by default, but return an attribution score of 0 Reviewed By: jjuncho Differential Revision: D71980024
sarahtranfb
force-pushed
the
export-D71980024
branch
from
5959648 to
4dd1e33
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71980024 |
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.
Summary:
For sparse features, the 0th dim is not always the batch size.
Currently for FeatureAblation, we skip the input tensor if there are no elements in it. For FeaturePermutation, we throw if we cannot permute across the "batch" (i.e. the tensor has only one element).
min_examples_per_batchparameterizes the above behavior. For FeaturePermutation, the default is set to 2, meaning we no longer throw by default, but return an attribution score of 0Differential Revision: D71980024