Perceptron
Implementation of the Perceptron Learning Algorithm in Python. The first graphic shows the perceptron binary classifier trained with 10 training points. Given this limited number of points, the decision boundary (the division between the blue and pink spaces) deviates from the original target function, which is shown as a bold white line.
When more training points are provided, such as the 100 training points in the second graphic, the original target function is better approximated by the decision boundary, leading to better performance from the classifier.
Perceptron trained with 10 training points - original target function in white
Key Features:
- Python-based implementation of perceptron binary classifier
- Includes methods for initializing, training and testing a perceptron
