torcs-rl

some experiments on the Torcs environment

installation guide

http://blog.naver.com/dnddnjs11/221074046378

example 1

Just see the state, action, reward of the environment Simple but important step in RL

example 2

Make simple actor-critic agent which can predict action but can't learn from interaction.

example 3

Make simple DDPG agent. See https:/yanpanlau/DDPG-Keras-Torcs I have made the code simpler.