Connect Four Game

A fully functional Connect Four game implementation in Python with AI opponent using Minimax algorithm with alpha-beta pruning.

Features

• Complete Connect Four gameplay with standard 6x7 board
• AI opponent using Minimax algorithm with alpha-beta pruning
• Heuristic evaluation for intelligent AI moves
• Human vs AI or Human vs Human modes
• Comprehensive test suite for board logic and AI behavior
• Command-line interface for easy gameplay

Project Structure

Connect_Four/
├── board.py          # Board class with game logic
├── ai.py             # MinimaxAI implementation
├── cli.py            # Command-line interface
├── verify.py         # Verification script
├── run_tests.py      # Test runner
├── tests/
│   ├── __init__.py
│   ├── test_board.py # Board tests
│   └── test_ai.py    # AI tests
└── README.md         # This file

Components

board.py

• Cell enum: Represents cell states (EMPTY, P1, P2)
• Board class: Game board with methods for:
  • Dropping pieces
  • Checking legal moves
  • Detecting winners (rows, columns, diagonals)
  • Board display and copying

ai.py

• MinimaxAI class: AI player using:
  • Minimax algorithm with alpha-beta pruning
  • Configurable search depth (default: 4)
  • Heuristic evaluation function
  • Strategic position scoring

cli.py

• Game loop for playing Connect Four
• Human input handling
• AI move execution
• Game state display

How to Play

Start the Game

python cli.py

Game Modes

• Human vs AI (default): Set vs_ai=True in cli.py
• Human vs Human: Set vs_ai=False in cli.py

Gameplay

1. The board displays with column numbers (0-6)
2. Player 1 (X) goes first
3. Enter column number to drop your piece
4. AI (O) responds automatically
5. First to connect 4 wins!

Board Display

  0 1 2 3 4 5 6
  -------------
  . . . . . . .
  . . . . . . .
  . . . . . . .
  . . . . . . .
  X . . . . . .
  X O . O . . .

• . = Empty cell
• X = Player 1
• O = Player 2 (AI)

Testing

Run Verification Script

python verify.py

This runs automated tests covering:

• Board creation
• Piece dropping
• AI decision making
• Winner detection
• AI blocking strategy

Run Test Suite

python run_tests.py

Or run individual test files:

python tests/test_board.py
python tests/test_ai.py

Implementation Details

AI Algorithm

The AI uses Minimax with Alpha-Beta Pruning:

• Searches game tree to configurable depth
• Evaluates positions using heuristics:
  • Center column preference (+3 points per piece)
  • Window evaluation (2, 3, 4 in a row)
  • Blocking opponent threats
  • Immediate win/loss detection

Winning Conditions

The game checks for 4 consecutive pieces in:

• Horizontal rows
• Vertical columns
• Diagonal lines (both directions)

Requirements

• Python 3.6+
• No external dependencies required