Initialize GradCAM mask summation with zeros. #22
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
From my read of the GradCAM paper, this matrix should be initialized with zeroes, so that cam contains only the weighted sum of the entries in outputs. Feel free to correct me here if I'm misinterpreting anything! [This doesn't affect the qualitative trend you see across the image in any of the generated GradCAM masks, since this is a constant shift at each pixel, but it does slightly perturb some of the numeric values.]
|
For VGG it doesn't matter, but for my network, this change was necessary to get meaningful results. It is indeed the correct approach. |
|
@jacobgil Friendly ping -- are you able to review this? |
|
I have a model with excellent accuracy an Roc but heat map generated by using grad cam is not reasonable? I used a architecture contains dome conv and batch normalizatio. |
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Outdated suggestions cannot be applied.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.
From my read of the GradCAM paper, this matrix should be initialized with zeroes, so that cam contains only the weighted sum of the entries in outputs. Feel free to correct me here if I'm misinterpreting anything!
[This doesn't affect the qualitative trend you see across the image in any of the generated GradCAM masks, since this is a constant shift at each pixel, but it does slightly perturb some of the numeric values.]