January 2020
Intermediate to advanced
346 pages
9h 8m
English
Since the architecture of the MLP is determined by the hidden layer configuration, let's explore how this configuration can be represented in our solution. The hidden layer configuration of the sklearn Multilayer Perceptron (https://scikit-learn.org/stable/modules/neural_networks_supervised.html) model is conveyed via the hidden_layer_sizes tuple, which is sent as a parameter to the model's constructor. By default, the value of this tuple is (100,), which means a single hidden layer of 100 nodes. If we wanted, for example, to configure the MLP with three hidden layers of 20 nodes each, this parameter's value would be (20, 20, 20). Before we implement our genetic algorithm-based optimizer for the ...
Read now
Unlock full access