Run OpenAI Gym environments on an external process or remote machine using gRPC.

Installation

Install Gym (not required if using Docker) and run:

pip install gymx

It is recommended to use a virtual environment.

Usage

Server

To start the server run:

python -m gymx

To use a different port run:

python -m gymx --port=54321

You can also run the server using Docker:

docker run -p 54321:54321 album/gymx

Client

Inside your application use:

from gymx import Env

env = Env('CartPole-v0')

To specify the server address use:

env = Env('CartPole-v0', address='localhost:54321')

API

• env.reset(): Reset the environment's state. Returns observation.
• env.step(action): Step the environment by one timestep. Returns observation, reward, done, next_episode. Unlike the original gym API, it automatically resets the environment when done and returns next episode's observation instead of info.