Neural Network Programming with TensorFlow

Third convolution layer

This layer takes the output of the second layer as the input. Let's look at the inputs going into the creation of this layer:

shape = [filter_size, filter_size, num_input_channels, num_filters] weights = new_weights(shape=shape)

layer_conv3, weights_conv3 = new_conv_layer(input=layer_conv2,                                            num_input_channels=num_filters2,filter_size=filter_size3,num_filters=num_filters3,use_pooling=True) 

print(layer_conv3)

The shape of layer_conv3 is as follows:

Tensor("Relu_2:0", shape=(?, 16, 16, 64), dtype=float32)