Guided backpropogation ReLU for Classification Layer

utkuozbulak / pytorch-cnn-visualizations

Pytorch implementation of convolutional neural network visualization techniques

MIT License

7.81k stars 1.49k forks source link

Closed yangerkun closed 3 years ago

yangerkun commented 3 years ago

It seems that only ReLU after conv layers are hooked. are there any reasons to not hook ReLU in the classifier module?

utkuozbulak commented 3 years ago

Hey,

There are only ReLUs as activation functions in the example model. Adapt it to your model based on your needs.