Enable attribute_to_layer_input both for layer activations and layer attribution #864
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
…attribution Summary: Currently, we cannot specify `attribute_to_layer_input` to LayerActivation. Since this flag has to be consistent both for attribution and layer activation, we can pass it through the constructor and use the same value for both layer activation and attribution. Removed unnecessary docs, e.g.: (a, b) (2 dimensions) There is no need for (2 dimensions). It is already clear that it has 2 dimensions. Also, simplified the computations based on the magnitude. The original one had an issue that if we had only one example and that example had positive TCAV score the magnitude was always 1.0 and that wasn't very useful. Differential Revision: D33908812 fbshipit-source-id: 1876e531f9bfa8e701bc678be2124703582fb3f4
|
This pull request was exported from Phabricator. Differential Revision: D33908812 |
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Outdated suggestions cannot be applied.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:
Currently, we cannot specify
attribute_to_layer_inputto LayerActivation. Since this flag has to be consistent both for attribution and layer activation, we can pass it through the constructor and use the same value for both layer activation and attribution.Removed unnecessary docs, e.g.: (a, b) (2 dimensions) There is no need for (2 dimensions). It is already clear that it has 2 dimensions.
Also, simplified the computations based on the magnitude. The original one had an issue that if we had only one example and that example had positive TCAV score the magnitude was always 1.0 and that wasn't very useful.
Differential Revision: D33908812