== DESCRIPTION ==

Demo for learning a double cart-pole task using Neural Fitted Q
Iteration.


== USAGE ==

To start the learning process, launch the script
./train
Note that learning requires the NFQControl module, which
is not included in the standard CLSquare bundle.
Task parameters can be adjusted in the config/ folder.

To replay selected policies from a previous run
instead, launch the script
./replay
The script features one randomly initialized network,
one that performs moderately well (~ episode 60) and
one that performs perfectly stable (~ episode 180).
The complete performance log of that run is included
in log/test1.stat (narrow initialization range) and
log/test2.stat (full initialization range).
