zealanl / rlgym-ppo

A vectorized implementation of PPO for use with RLGym

RLGym-PPO

A vectorized implementation of PPO for use with RLGym.

INSTALLATION

1. install RLGym-sim.
2. If you would like to use a GPU install PyTorch with CUDA
3. Install this project via pip install git+https://github.com/AechPro/rlgym-ppo

USAGE

Simply import the learner with from rlgym_ppo import Learner, pass it a function that will return an RLGym environment and run the learning algorithm. A simple example follows:

from rlgym_ppo import Learner

def my_rlgym_function():
    import rlgym_sim
    return rlgym_sim.make()

learner = Learner(my_rlgym_env_function)
learner.learn()

Note that users must implement a function to configure Rocket League (or RocketSim) in RLGym that returns an RLGym environment. See the example.py file for an example of writing such a function.