plot_decision_boundary.py

plot_decision_boundary.py

def plot_decision_boundary(perceptron, X, y, title="Perceptron Decision Boundary"):
    """
    Plots the decision boundary of a trained perceptron.

    Args:
        perceptron (Perceptron): The trained perceptron model.
        X (numpy.ndarray): The input data points (N x 2).
        y (numpy.ndarray): The corresponding labels (N,).
        title (str, optional): The title of the plot. Defaults to "Perceptron Decision Boundary".
    """

    plt.figure()

    # Plot data points
    plt.scatter(X[y == 1, 0], X[y == 1, 1], c='blue', marker='o', label='Label 1')
    plt.scatter(X[y == -1, 0], X[y == -1, 1], c='red', marker='^', label='Label -1')

    # Calculate decision boundary
    b, w1, w2 = perceptron.w_  # Extract weights and bias

    # Create a line that represents the decision boundary
    x1_min, x1_max = X[:, 0].min() - 1, X[:, 0].max() + 1
    x2_min, x2_max = X[:, 1].min() - 1, X[:, 1].max() + 1

    x1_line = np.linspace(x1_min, x1_max, 100)
    x2_line = -(w1 * x1_line + b) / w2  # Equation of the line: w1*x1 + w2*x2 + b = 0

    # Plot the decision boundary
    plt.plot(x1_line, x2_line, 'r:', label='Decision Boundary')

    plt.xlabel('x1')
    plt.ylabel('x2')
    plt.title(title)
    plt.legend()
    plt.grid(True)
    plt.show()

# Example
epochs = 100
lr = 0.1

# Loading datasets
X_train, y_train = load_dataset("train_dataset1")
X_test,  y_test = load_dataset("test_dataset1")

# Training Perceptron
perceptron = Perceptron(features_count=X_train[0].shape[0], learning_rate=lr, epochs=epochs)
perceptron.fit(X_train, y_train)

plot_decision_boundary(perceptron, X_train, y_train, "Dataset 1 - Training Scatter Plot")