Perceptron

Rosenblatt's Perceptron in Python

Requirements

• python3
• numpy
• unittest

Run tests

python3 perceptron_test.py

How to use

You first need to prepare your data. In this example the Perceptron will learn the AND operation.

target_values = np.array([0, 0, 0, 1])
source_inputs = np.array(([0, 0], [0, 1], [1, 0], [1, 1]))

You then need to create a Perceptron with a given number of input and an activation function

p = Perceptron(2, binary_step)

note that you can specify the maximum number of iterations, the learning constant and the error threshold.

You can now train the Perceptron and make it predict an output given an input:

perceptron.train(source_inputs, target_values)

output = perceptron.predict([1, 1])
print(output)

The output should be 1.