coreylynch / async-rl

Tensorflow + Keras + OpenAI Gym implementation of 1-step Q Learning from "Asynchronous Methods for Deep Reinforcement Learning"
MIT License
1.01k stars 174 forks source link

Add python 3 support #28

Closed expenses closed 4 years ago

expenses commented 6 years ago

Rendering the screens doesn't look quite right, but it runs at least.