The idea is to train an agent to navigate, and collect points — these are bananas!, who doesn't like them? 😋— in a large square world using Deep Q-Network.
A reward of +1 is provided for collecting a yellow banana, and a reward of -1 is provided for collecting a purple banana. Thus, the goal of the agent is to collect as many yellow bananas as possible while avoiding the purple ones.
The state space has 37 dimensions and contains the agent's velocity, along with ray-based perception of objects around agent's forward direction. Given this information, the agent has to learn how to best select actions. Four discrete actions are available, corresponding to:
0- move forward.1- move backward.2- turn left.3- turn right.
The task is episodic, and in order to solve the environment, the agent must get an average score of +13 over 100 consecutive episodes.
- Untrained Agent:
- Trained Agent:
For more technical details about this project, algorithms, training, results, comparison, etc... please go to Report
-
Download the environment from one of the links below. You need only select the environment that matches your operating system:
- Linux: click here
- Mac OSX: click here
- Windows (32-bit): click here
- Windows (64-bit): click here
(For Windows users) Check out this link if you need help with determining if your computer is running a 32-bit version or 64-bit version of the Windows operating system.
(For AWS) If you'd like to train the agent on AWS (and have not enabled a virtual screen), then please use this link to obtain the environment.
-
Place the file where your Navigation.ipynb is, and unzip (or decompress) the file.
-
Other important packages to install:
pip install numpy
pip install torch
pip install unityagents
Follow the instructions in Navigation.ipynb to get started with training your own agent!.
Sometimes Github cannot load Jupyter notebooks. If that's the case, click here to visualize it using nbviewer.