Perceptron

A simple, generic implementation of a Perceptron in Go (Golang), supporting multi-dimensional datasets and optional training history recording.

Disclaimer
This project was created only for educational purposes.
There are no guarantees regarding correctness, performance, or suitability for any real-world application.

Overview

This package implements a basic linear binary classifier (perceptron) for numeric datasets.
It supports:

• Training on multi-dimensional features.
  • 2D Example:
  • 3D Example:
• Training history tracking (optional) — including weights, bias, and mean error after each epoch.

Features

• Trainable perceptron with a customizable learning rate and epochs.
• Predict new data points after training.
• Access complete training history for analysis or visualization.
• Lightweight and dependency-free (except for golang.org/x/exp/constraints).