What the network “thinks” is the best image for the CNN model ? (Class Maximization tutorial)

What If we asked our deep neural network to draw it’s best image for a trained model ?

What it will draw ? What is the optimized image for each model category ?


We can discover that using the class maximization method on the Vgg16 model.


You can find more similar tutorials in my blog posts page here : https://eranfeit.net/blog/

You can find the link for the video tutorial here: &list=UULFTiWJJhaH6BviSWKLJUM9sg



Enjoy

Eran