DEPRECATED

This package is deprecated.

OpenAIGymAPI.jl

Wrapper for the OpenAI Gym API. For the official JuliaML OpenAIGym wrapper please see OpenAIGym.jl.

Package Status	Package Evaluator	Build Status

Introduction

OpenAI Gym is a open-source Python toolkit for developing and comparing reinforcement learning algorithms. This Julia package is a wrapper for the OpenAI Gym API, and enables access to an ever-growing variety of environments.

Installation

This package is registered in METADATA.jl and can be installed as usual

Pkg.add("OpenAIGymAPI")
using OpenAIGymAPI

If you encounter a clear bug, please file a minimal reproducible example on Github.

Setting up the server

To download the code and install the requirements, you can run the following shell commands:

git clone https://github.com/openai/gym-http-api
cd gym-http-api
pip install -r requirements.txt

This code is intended to be run locally by a single user. The server runs in python.

To start the server from the command line, run this:

python gym_http_server.py

For more details, please see here: https://github.com/openai/gym-http-api.

Overview

using OpenAIGymAPI

remote_base = "http://127.0.0.1:5000"
client = GymClient(remote_base)
print(client)

# Create environment
env_id = "CartPole-v0"
instance_id = env_create(client, env_id)
print(instance_id)

# List all environments
all_envs = env_list_all(client)
print(all_envs)

# Set up agent
action_space_info = env_action_space_info(client, instance_id)
print(action_space_info)
agent = action_space_info["n"] # perform same action every time

# Run experiment, with monitor
outdir = "/tmp/random-agent-results"
env_monitor_start(client, instance_id, outdir, force = true, resume = false)

episode_count = 100
max_steps = 200
for i in 1:episode_count
  ob = env_reset(client, instance_id)
  done = false
  j = 1
  while j <= 200 && !done
    action = env_action_space_sample(client, instance_id)
    results = env_step(client, instance_id, action, render = true)
    done = results["done"]
    j = j + 1
  end
end

# Dump result info to disk
env_monitor_close(client, instance_id)

License

This code is free to use under the terms of the MIT license.

Acknowledgements

The original author of OpenAIGymAPI is @Paul Hendricks.