pytorch-distributed-reinforcement-learning

Architecture on Distributed Multi Machines

Algo

PPO

PPO-Continuous

IMPALA

V-MPO

SAC

SAC-Continuous

Caution!

Discrete Learning environment is configured to CartPole-v1.

Continuous Learning environment is configured to MountainCarContinuous-v0.

You should check machines.json, parameters.json for architecture and training parameters.

How to run

python run.py