April 2019
Intermediate to advanced
212 pages
5h 34m
English
A simple neural network, such as a classifier, might consist of an input layer, a hidden layer, and an output layer. Each layer has a set of weights associated with it.
The values in the input layer of the network propagate through the network. The basis for calculating the node values of a new layer is by taking the dot product of the input and the network weights:

The output is a matrix of prediction values. As with any machine learning classifier, the range of values is based on the number of classes. For a binary classifier, for example, the output values will be 0 or 1.
The goal of the network is to ...
Read now
Unlock full access