/deep-reinforcement-learning_DDQN_PPO_HER

MLP-framework (pure numpy) and DDQN-framework for OpenAI's Gym games. +test code for PPO added. +Hindsight Experience Replay(HER) bitflip-DQN example. +prioritized replay.

Primary LanguageJupyter Notebook

Deep Reinforcement Learning in Games

MLP framework and DDQN framework for OpenAI's gym game environments. - written in numpy
Take a look at the Examples in the Documentation.

Installation

The mlp_framework.py shouble able to run on nearly anything that can interpret python3 (and numpy).
The ddqn_framework.py additionally uses OpenAI's gym
Download the repo and run the jupyter notebook.