Skip to content

Latest commit

 

History

History
50 lines (34 loc) · 1.53 KB

README_MINIMAL.md

File metadata and controls

50 lines (34 loc) · 1.53 KB

Marlo

marLo : Multi Agent Reinforcement Learning using MalmÖ

NOTE : THIS IS A WORK IN PROGRESS. AND NOT READY FOR THE FIRST RELEASE YET

Marlo requires Python3 and Java8 to be installed. On Windows Python3 MUST be the 64bit version.

Install malmo

pip3 install malmo

Alternatively install malmo from GitHub

If the above fails then you may need to install malmo manually: Please see Malmo GitHub on how to install Malmo from a binary release or building from source. On Linux and MacOS, You will need to include the MalmoPython.so library on your PYTHONPATH. Mincraft can then be launched from the malmo/Minecraft directory using the launchClient.sh or launchClient.bat script with the -port argument specifying the port.

Install marlo from GitHub

git clone https://github.com/spMohanty/marlo
cd marlo
python setup.py install

Running the multi agent examples

The multi-agent examples can be run from the marlo directory.

Assuming malmo was installed with pip, you can start Minecraft on ports 1000 and 10001 by running the following in a cmd terminal or shell:

python3 two_agent_minecraft_launch.py

Once two minecraft windows appear, run the multi agent example with:

python3 -m examples.multiagentexample --turn_based --rollouts 10  --mission_file mob_chase.xml

or for the full Chainer RL experience:

python3 -m examples.multiagent_chainer_test_DQN --turn_based --rollouts 10  --mission_file mob_chase.xml