/LeducRebel

Implementation of the Rebel algorithm on the Ludec Hold em Poker game in RLCard's toolkit environment

Primary LanguagePythonMIT LicenseMIT

CS 172B Project

Attempt to implement the Rebel algorithm to train an agent to play Ludec Hold em Poker.

Installation guide - From the root of the project, change into the examples directory - run the command in terminal: 'pip install -r requirements.txt'

Running experiments - From the root of the project, change into the examples directory - run any of the following scripts to generate experimental results run_cfr.py run_random.py run_rebel.py run_rl.py

Visualizing results - From the root of the project, run the plot.ipynb notebook.

Refer to https://github.com/datamllab/rlcard