DuelQSimple: Double Duelling Deep Q Learning

TODO reference the original paper

Description

This file serves as an concrete example on how to implement a baseline, even more concretely than the “do nothing” baseline. Don’t expect to obtain state of the art method with this simple method however.

An example to train this model is available in the train function Example-duelqsimple.

Warning

This baseline recodes entire the RL training procedure. You can use it if you want to have a deeper look at Deep Q Learning algorithm and a possible (non optimized, slow, etc. implementation ).

For a much better implementation, you can reuse the code of l2rpn_baselines.PPO_RLLIB or the l2rpn_baselines.PPO_SB3 baseline.

Exported class

You can use this class with:

from l2rpn_baselines.DuelQSimple import train, evaluate, DuelQSimple

Other non exported class

These classes need to be imported, if you want to import them with (non exhaustive list):

from l2rpn_baselines.DuelQSimple.duelQ_NN import DuelQ_NN
from l2rpn_baselines.DuelQSimple.duelQ_NN import DuelQ_NNParam