5l1v3r1 / ddqn_banana_picking_game

Project Overview

This my final project, in this project I trained an agent to pick up yellow bananas that are scattered throughout a square world. The agent must learn to navigate in its environment to reach yellow bananas (reward of +1 each) while avoiding to collect blue bananas (reward of -1 each).

Instructions

Follow the instructions in final_project.ipynb to train an agent from scratch or simply watch a pretrained agent

Settings

To set up your python environment to run the code, follow the instructions below.

1. Create (and activate) a new environment with Python 3.6.

   o Linux or Mac:

        conda create --name drlnd python=3.6
   
        source activate drlnd
   

   o Windows:

        conda create --name drlnd python=3.6 
   
        activate drlnd
   

2. terminal:

      pip install ipykernel
   

3. Create an IPython kernel for the drlnd environment.

      python -m ipykernel install --user --name drlnd --display-name "drlnd"
   

4. Before running code in a notebook, change the kernel to match the ‘drlnd’ environment by using the drop-down Kernel menu.

5. terminal:

      pip install unityagents,torch,scipy
   

6. I have download the unity file in the package ,if you want to update Banana.unity, go to this github source https://github.com/udacity/deep-reinforcement-learning/tree/master/p1_navigation

7. Follow the instructions in Navigation.ipynb to test the environment.