Example of Perceptron Logic Gates AND, OR and XOR to build Artificial Neural Networks.

perceptron_logic_gates.py

from sklearn.linear_model import Perceptron
import matplotlib.pyplot as plt
import numpy as np
from itertools import product

data = [[0, 0], [0, 1], [1, 0], [1, 1]]

# Set labels basing on gate type (AND, OR)
# Multiple perceptrons can solve XOR gates
labels = [0, 0, 0, 1]
x = [point[0] for point in data]
y = [point[1] for point in data]
plt.scatter(x, y, c=labels)

# Create Perceptron
classifier = Perceptron(max_iter=40)
classifier.fit(data, labels)
score = classifier.score(data, labels)

# Defining the decision points (100 between 0 and 1)
x_values = np.linspace(0, 1, 100)
y_values = np.linspace(0, 1, 100)

# Calculate every single combination
point_grid = list(product(x_values, y_values))

# Get distances in absolute values
distances = classifier.decision_function(point_grid)
abs_distances = [abs(d) for d in distances]

# Create 2D array (matrix)
distance_matrix = np.reshape(abs_distances, (100, 100))

# Draw heat map
heatmap = plt.pcolormesh(x_values, y_values, distance_matrix)
plt.colorbar(heatmap)

# Show The Matrix
plt.show()