https://github.com/s2t2/tic-tac-toe-py
a basic adversarial game
https://github.com/s2t2/tic-tac-toe-py
ai alpha-beta-pruning game minimax-algorithm tic-tac-toe
Last synced: 8 days ago
JSON representation
a basic adversarial game
- Host: GitHub
- URL: https://github.com/s2t2/tic-tac-toe-py
- Owner: s2t2
- License: other
- Created: 2021-12-21T23:14:52.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2024-07-25T21:25:14.000Z (over 1 year ago)
- Last Synced: 2024-07-26T00:03:14.044Z (over 1 year ago)
- Topics: ai, alpha-beta-pruning, game, minimax-algorithm, tic-tac-toe
- Language: Python
- Homepage:
- Size: 258 KB
- Stars: 30
- Watchers: 3
- Forks: 175
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Tic Tac Toe
An adversarial game.
[](https://codeclimate.com/github/s2t2/tic-tac-toe-py/maintainability)
## Prerequisites
+ Anaconda 3.7+
+ Python 3.7+
+ Pip
## Setup
Optionally fork this [remote repository](https://github.com/s2t2/tic-tac-toe-py), to create a copy under your own control. Then "clone" or download the remote repository (or your forked copy) onto your local computer, for example to your Desktop. Then navigate to wherever you downloaded the repo:
```sh
cd ~/Desktop/tic-tac-toe-py
```
Create a virtual environment:
```sh
conda create -n tictactoe-env python=3.11
```
Activate the virtual environment:
```sh
conda activate tictactoe-env
```
Install package dependencies within the virtual environment:
```sh
pip install -r requirements.txt
```
## Player Types
When you play games, you'll be able to select any of the following combinations of players:
player type(s) | description
--- | ---
`HUMAN` | A human player who will input their selections.
`COMPUTER-EASY` or `RANDOM` | A computer player which makes random selections. Easy to beat.
`COMPUTER-HARD` or `MINIMAX` | A computer player which thinks ahead to make optimal selections. Uses the "minimax" algorithm to simulate moves and evaluate all possible game states. Impossible to beat.
`MINIMAX-AB` | A much faster version of the hard computer player. Uses "alpha-beta" pruning to skip evaluations of unnecessary game states.
## Usage
### Game Play
Play a game (human vs human, human vs computer, computer vs computer):
```sh
python -m app.game
```
### Game Simulation
Play many games (computer vs computer), saves results to CSV file in "data" directory. Optionally pass the game count and/or player strategies as environment variables:
```sh
python -m app.jobs.play_games
# alternatively:
GAME_COUNT=3 python -m app.jobs.play_games
# alternatively:
X_STRATEGY="COMPUTER-HARD" O_STRATEGY="COMPUTER-EASY" GAME_COUNT=100 python -m app.jobs.play_games
```
## Testing
Run automated tests, to know whether the app is working as expected:
```sh
pytest
```
## Demo
Here is a demonstration of gameplay between computer players:
```
SELECT X PLAYER TYPE ('HUMAN' / 'COMPUTER-EASY' / 'COMPUTER-HARD'): COMPUTER-HARD
SELECT O PLAYER TYPE ('HUMAN' / 'COMPUTER-EASY' / 'COMPUTER-HARD'): COMPUTER-HARD
Would you like to use a pre-saved game state? (Y/N): n
A B C
1 | |
-----------
2 | |
-----------
3 | |
PLAYER X THINKING...
A B C
1 | |
-----------
2 | |
-----------
3 X | |
PLAYER O THINKING...
... -3
... -3
... -3
... -3
... 0
... -3
... -3
... -3
A B C
1 | |
-----------
2 | O |
-----------
3 X | |
PLAYER X THINKING...
... 0
... 0
... 0
... 0
... 0
... 0
... 0
A B C
1 X | |
-----------
2 | O |
-----------
3 X | |
PLAYER O THINKING...
... -5
... -5
... 0
... -5
... -5
... -5
A B C
1 X | |
-----------
2 O | O |
-----------
3 X | |
PLAYER X THINKING...
... -4
... -4
... 0
... -4
... -4
A B C
1 X | |
-----------
2 O | O | X
-----------
3 X | |
PLAYER O THINKING...
... 0
... 0
... 0
... 0
A B C
1 X | O |
-----------
2 O | O | X
-----------
3 X | |
PLAYER X THINKING...
... -2
... 0
... -2
A B C
1 X | O |
-----------
2 O | O | X
-----------
3 X | X |
PLAYER O THINKING...
... -1
... 0
A B C
1 X | O |
-----------
2 O | O | X
-----------
3 X | X | O
PLAYER X THINKING...
... 0
A B C
1 X | O | X
-----------
2 O | O | X
-----------
3 X | X | O
{'winner': None, 'reason': 'NO_MORE_SQUARES', 'message': 'TIE GAME'}
```
## [License](/LICENSE.md)