Implementation of layer-wise relevance propagation on a convolutional neural network trained on the MNIST handwritten digit dataset.
Example of heat maps for select layers for a sample input image of 7:
The corresponing relevance scores for every output class for the smaple input image of 7:
Example of heat maps for select layers for a sample input image of 8:
The corresponing relevance scores for every output class for the smaple input image of 8:
\
-
Notifications
You must be signed in to change notification settings - Fork 1
seanseungbeomlee/PyTorch-CNN-LRP
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Implementation of layer-wise relevance propagation on a convolutional neural network trained on the MNIST handwritten digit dataset.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published