January 2018
Beginner to intermediate
284 pages
8h 35m
English
The first thing in backpropagation is to calculate the errors from forward propagation for your target value. The input provides y as a test for the accuracy of the network’s output, so we compute the following vector:
This is written in code as follows:
# define error, which is the difference between the activation function output from the last layer and the labelerror = tf.sub(a_2, y)
Read now
Unlock full access