PokerBot is an interactive poker app allowing human versus AI play. Leveraging the Pygame and NumPy libraries, PokerBot features an easy to use GUI and a reinforcement-learning opponent that improves as it plays.
Vivien Chen, Harris Davidson, Ava Lakmazaheri, Emma Westerhoff, Harrison Young
We would like to thank Denny Britz for his reinforcement learning code, which we used as reference.
All the libraries we used in the main code, as well as in our test code and old versions, are included in our requirements.txt file. You can install them by running:
- pip install -r requirements.txt
Alternatively, you can choose which libraries to install as necessary by typing the following into your command line:
- pip install pygame
- pip install gym
- pip install pprint
- pip install matplotlib
- pip install numpy
- pip install flask
The PokerBot's machine learning data is loaded from, updated to, and saved in sa_cache.txt.
To install the PokerBot, simply clone the GitHub repository and run python control.py in the command line. The game will immediately begin.
For more information about the development of PokerBot, please visit our website: https://hackathon64.wixsite.com/pokerbot or take a look at our poster.