trabelsim / MDP

Value iteration and QLearning

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

MDP Solving

Markov Decision Process ValueIteration algorithm

Configuration

The provided directory already contains two loaded worlds. In case of testing additional worlds it is necessary to copy the world data files into the current directory. Important! The file has to be .data format.

Usage

/.mdp SOURCE_WORLD_DATA_FILE [gamma] [explore]

Example 1

/.mdp MDPRL_world1.data

Example 2

/.mdp MDPRL_world1.data 0.90 

License

MIT

About

Value iteration and QLearning


Languages

Language:Python 100.0%