Project: AlphaZero-Inspired Game Learning: Faster Training by Using MCTS Only at Test Time 
Group Number: 54

Description: This project implements Othello and ConnectFour, with AI vs Human modes, AI vs Edax comparisons (for Othello), and performance analyses using MCTS at test time.

Prerequisites:
1.Ensure you have Python 3.x installed on your system.
2.Install required dependencies:
  pip install numpy
  pip install matplotlib

Steps to Run Othello (Human VS Our AI Mode):
3.Run the following command in the terminal:
  cd Othello
  python othello.py
  
  Press *Play Game* button to start the game.
  Press *New Game* button to start the game and then start playing.

  Directory Structure:
  ├── controller
  │   ├── othello_controller.py
  ├── model
  │   ├── othello_model.py
  ├── outputs
  |   ├── aivshumanothelloGame1.jpg
  |   ├── aivshumanothelloGame2.jpg
  |── views
  |   ├── othello_view.py
  ├── othello.py  (main script)
  othello_agent.pkl  (trained model)

Steps to Run ConnectFour (Human VS Our AI Mode):
3.Run the following command in the terminal:
  cd ConnectFour
  python connect_four.py

  Directory Structure:
  ├── outputs
  |   ├── aivshumanc4game1.png
  |   ├── aivshumanc4game2.png
  |   |── aivshumanc4game3.png
  |── connect_four.py (main script)
  |── ConnectFourModel.pkl (trained model)

Steps to Analyse and Plot the Results of ConnectFour Games:
3.Run the following command in the terminal:
  cd ConnectFourAnalysis
  python tournament.py

It runs the game and creates a brief analysis of MCTS Vs TCL-wrap, MCTS Vs TCL-base, and MCTS Vs AB-DL Agent
  Directory Structure:
  ├── graph
  |   ├── analysis.png
  ├── model.pkl (trained model)
  |── tournament.py (main script)


Steps to Run Othello (AI VS Edax Mode):
3.Run the following command in the terminal:
  cd OurAIvsEdax
  python edaxvsothello.py

  Press *New Game* button to start the game.
  Then both AIs will play against each other.

  Directory Structure:
  ├── controller
  │   ├── edax_wrapper.py
  │   ├── edaxvsothello_controller.py
  ├── model
  │   ├── othello_model.py 
  │   ├── td_agent.py 
  │   ├── edaxvsothello_model.py
  ├── outputs
  |   ├── graphs
  |   |   ├── win_rates_level_5.png
  |   |   ├── win_rates_level_6.png
  |   |   ├── win_rates_level_8.png
  |   ├── edaxvsaiLevel5.png
  |   ├── edaxvsaiLevel8.mp4
  |   ├── edaxvsaiLevel8.png
  |── views
  |   ├── othello_game.py
  |   ├── edaxvsothello_view.py
  |── othello_agent.pkl (trained model)
  ├── edaxvsothello.py (main script)
  edax.exe (edax version 4.6)

To generate graph for analysing the win rates of Edax at different levels vs Our AI:
  python edaxlevelwisegraph.py

Additional Notes:
1. Outputs and visualizations are stored in the respective outputs and views folders.
2. The implementation leverages MCTS at test time to enhance training efficiency, as per the research paper.