How is feature extraction done from by using pre-activations last CNN layer in VGG-19

I am trying to extract features using the pre-activations of last CNN layer in VGG-19.Is it just extracting features before the ReLU layer or do I have to make any changes in the CNN model…?

You could use forward hooks to get the activations without changing the model as described here.

1 Like

Thank you…It worked out…