search

nadavbh12 / Retro-Learning-Environment

The Retro Learning Environment (RLE) -- a learning framework for AI
Other
183 stars 40 forks source link

Help Needed, Resulted Trained agent? #6

Closed misterabdullahAziz closed 7 years ago

misterabdullahAziz commented 7 years ago
  1. how to add new game?
  2. where is final result of trained agent?
  3. How to switch dqn with ddqn d-ddqn?

I am all new in this So your help will be highly appreciated. Thanks, Looking forward.

nadavbh12 commented 7 years ago

Hi,

  1. We created a wiki page with instructions for adding a new game.

  2. and 3. This repository is an environment on which you can train a variety of agents. The only agents provided here are sample agents in C++ and Python. You can check out our fork of deep_q_rl which runs the basic DQN on top of RLE in Python. Porting other implementation to use RLE rather than ALE should be fairly simple as the interface is mostly the same.