SuperMario_DeepQ

A Deep Q RL agent based on "Playing Atari with Deep Reinforcement Learning" By Mnih et al to solve Super Mario Game

Setup

In a python 3.7+ environment, run:

conda create --name mario python=3.9
conda activate mario
conda install pytorch
pip install -r requirements.txt

Training

To train the model, run:

python main_DQ.py

The training hyperparameters are stored in ./configs/config_DeepQ.yaml

Run Pretrained Model

A pretrained RL agent model in available in ./checkpoints. To play mario with this model, run:

python play_DQ.py