visualize activation at different layers in CNN in Keras

keras-activation-visualization-cnn.md

First get list of all the layers (after activation).

from keras.models import Model
layer_outputs = [layer.output for layer in model.layers]

Predict image and get corresponding activation in each layer

activation_model = Model(inputs=model.input, outputs=layer_outputs)
activations = activation_model.predict(image)

Plot Activation

def display_activation(activations, col_size, row_size, act_index): 
    activation = activations[act_index]
    activation_index=0
    fig, ax = plt.subplots(row_size, col_size, figsize=(row_size*2.5,col_size*1.5))
    for row in range(0,row_size):
        for col in range(0,col_size):
            ax[row][col].imshow(activation[0, :, :, activation_index], cmap='gray')
            activation_index += 1

Here, act_index represent the layer whose activation we are going to visualize.

Display

Plot activation of layer at index 5.

display_activation(activations, 4, 4, 5)