README.md

Reinforcement Learning and VR

Files Included

vr

• gym-vr: a pip package for VR gym environment
• dqn: a deep Q-Learn model
• netDelay: source code for generating ./Program
• FYI
  • log: set of logs extracted from ouput of MI analyzers
  • set of analyzers: the analyzer scripts I am using

Usage

First we need to install VR training environment by

cd VR_REIN
pip install -e .

This environment strictly follows package "gym"'s requirement, which gives observation, reward for each action in action space.

Then we start a sample server which listens to 127.0.0.1 with port number 9999 by

cd netDelay/server
./Program

Then we run the model by

cd dqn
python dqn.py