BipedalWalker-v2

OpenAI game.

The repo is under development, there are some bugs in some of the scripts.

Methods

1. Deep-Q-Learning Networks based on Reinforcement-Learning
   
2. Experiment with Recurrent Deterministic Policy Gradient Methods
3. General research on reinforcement learning algorithms
4. Actor-Critic Networks
5. A2C Algorithms
6. PPO

Instructions

1. In order to run both scripts you need to install openAI-gym lib.

2. For gym install follow the next steps:

    $ pip install --upgrade pip
    $ git clone https://github.com/openai/gym.git
    $ cd gym
    $ pip install -e '.[box2d]'

    # test that it worked 
    $ python >>> import gym >>> gym.make('BipedalWalker-v2')

    # The above instructions are for linux systems. 
    # We do not use proprietary software so we do not have instructions for Windows or MacOS.  

You can visit openAI gym in order to inspect various environments.

3. Other dependencies : Keras, matplotlib, numpy, random, tensorflow

4. Do not use any Python-IDE to run the code if you want to see the game-simulation (environment-render). Run the code in a console or a terminal by typping: $ python script-name.py