How to do it...

  1. First, we start with loading the libraries, as follows:
import osimport globimport tensorflow as tf
  1. Because our task is slightly more complicated than outputting the predicted class, we need to define a function that extracts the values from different layers:
def extract_layers(vgg_layer3_out, vgg_layer4_out, vgg_layer7_out, n_classes):    decode_layer1_preskip0 = tf.layers.conv2d_transpose(vgg_layer7_out, 512, (2, 2), (2, 2), name='decode_layer1_preskip0')    decode_layer1_preskip1 = tf.layers.conv2d(vgg_layer4_out, 512, (1, 1), (1, 1), name='decode_layer1_preskip1')    decode_layer1_out = tf.add(decode_layer1_preskip0, decode_layer1_preskip1, name='decode_layer1_out')    decode_layer2_preskip0 = tf.layers.conv2d_transpose(decode_layer1_out, ...

Get Python Deep Learning Cookbook now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.