-This project uses the OpenAI Gym. https://gym.openai.com/
-It also uses the nes-py to use the controller to play the game.
-Install the dependencies from the requirements.txt.
pip3 install -r requirements.txt
-Do 2 types of preprocessing before applying reinforcement learning
-converting the rgb frames to gray
-frame stacking
-Use the PPO algorithm
-To view the tensorboard statistics run from logs/PPO_20
tensorboard --logdir=.