How to optimize networking using optimization algorithms in PyBrain

Recipe Objective - How to optimize networking using optimization algorithms in PyBrain?

Pybrain provides a GA optimization algorithm to optimize a network.

For more related projects -

https://www.projectpro.io/projects/data-science-projects/deep-learning-projects
https://www.projectpro.io/projects/data-science-projects/tensorflow-projects

Let's try to build and optimize a network of AND gates using the GA optimization algorithm -

# Importing libraries
from pybrain.datasets.classification import ClassificationDataSet
from pybrain.optimization.populationbased.ga import GA
from pybrain.tools.shortcuts import buildNetwork

# create AND dataset
and_dataset = ClassificationDataSet(2)

# Adding sample to and_dataset
and_dataset.addSample([0., 0.], [0.])
and_dataset.addSample([0., 1.], [0.])
and_dataset.addSample([1., 0.], [0.])
and_dataset.addSample([1., 1.], [1.])

# Setting target field
and_dataset.setField('class', [[0.],[0.],[0.],[1.]])

# Building network with 2 input layers, 3 hidden layers and 1 output layer
build_network = buildNetwork(2, 3, 1)

# GA optimization algorithm
ga_optimization = GA(and_dataset.evaluateModuleMSE, build_network, minimize=True)

# 50 iterations for learning
for i in range(50):
   build_network = ga_optimization.learn(0)[0]

# Activating network by passing some input
print(build_network.activate([1,1]))


Output -
[1.339035]

In this way, we can optimize a network in pybrain.

​