Gym-MicroRTS

This repo contains the source code for the gym wrapper of MicroRTS authored by Santiago Ontañón.

Get Started

Go to http://microrts.s3-website-us-east-1.amazonaws.com/microrts/artifacts/ to download the latest version of MicroRTS build. For example:

$ mkdir ~/microrts && \
wget -O ~/microrts/microrts.zip http://microrts.s3.amazonaws.com/microrts/artifacts/201910011635.microrts.zip && \
unzip ~/microrts/microrts.zip -d ~/microrts/ && \
rm ~/microrts/microrts.zip
~/microrts

Then clone the repo

$ git clone https://github.com/vwxyzjn/gym-microrts.git && \
cd gym-microrts && \
pip install dacite && \
pip install -e .
$ python3 hello_world.py

And run either the hello_world.py referenced above or the following file

import gym
import gym_microrts

# The prefix `Eval` renders the gmae.
env = gym.make("EvalMicrortsGlobalAgentsProd-v0")
# Alternatively, try headless mode at
# env = gym.make("MicrortsGlobalAgentsProd-v0")
env.reset()
for _ in range(10000):
    env.step(env.action_space.sample())
env.close()

Performances of agents on different map sizes

https://app.wandb.ai/costa-huang/MicrortsRL