Be sure you have python 3.6 installed.
Run install.sh
. This creates the python environment mgenv
with the necessary requirements to run the application.
To install as a python package, run pip install -e .
You can use the simulator as a Gym environment, see examples/env_demo.py
Be sure you are using the python environment mgenv
(source menv/bin/activate
).
Get help::
python -m microgridRLsimulator -h
Run the application::
python -m microgridRLsimulator --from_date 20150101T00:00:00 --to_date 20150102T00:00:00 --agent DQN --agent_options agent_options.json case1
This runs the case1 examples located in folder examples/data
using a DQNAgent, that is a agent that uses Deep Q learning.
You can generate this documentation yourself:
::
cd <to the root of the project>
cd docs; make html; cd ..
The html doc is in _build/html