ML Practice Repository

This repository contains machine learning practice projects covering different areas of ML, including reinforcement learning and regression tasks.

Projects

1. Connect 4 Reinforcement Learning (connect_4_RL/)

A reinforcement learning implementation for playing Connect 4. This project demonstrates how to apply RL techniques to a classic board game.

Key Components:

• Game engine with win detection and move validation
• RL environment wrapper compatible with standard RL libraries
• Terminal and Pygame interfaces for human play
• Jupyter notebook for experimentation and development

Technologies: Python, NumPy, Pygame (optional)

2. Melbourne House Price Prediction (house_pricing/)

A regression project to predict house prices in Melbourne using machine learning models. This project showcases end-to-end ML pipeline including data preprocessing, model training, and evaluation.

Key Components:

• Data loading and preprocessing pipeline
• Multiple model implementations (training and evaluation)
• Experiment tracking and logging
• Performance metrics calculation

Technologies: Python, Pandas, Scikit-learn, Jupyter

Setup

Prerequisites

• Python 3.8 or higher
• pip or conda for package management

Installation

Each project has its own dependencies. Navigate to the project directory and install:

Connect 4 RL

cd connect_4_RL
pip install -e .