Solving a variation of the Unity tennis environment using deep reinforcement learning. This repo contains a project to meet requirements for the Udacity Reinforcement Learning Nanodegree.
This is the first project in Udacity's Deep RL nanodegree. In this project we work with the Tennis environment.
In this environment, two agents control rackets to bounce a ball over a net. If an agent hits the ball over the net, it receives a reward of +0.1. If an agent lets a ball hit the ground or hits the ball out of bounds, it receives a reward of -0.01. Thus, the goal of each agent is to keep the ball in play.
The observation space consists of 8 variables corresponding to the position and velocity of the ball and racket. Each agent receives its own, local observation. Two continuous actions are available, corresponding to movement toward (or away from) the net, and jumping.
The environment is solved when the agent achieves an average score of +0.5 (over 100 consecutive episodes, after taking the maximum over both agents).
-
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 "headless" version of the environment. You will not be able to watch the agent without enabling a virtual screen, but you will be able to train the agent. (To watch the agent, you should follow the instructions to enable a virtual screen, and then download the environment for the Linux operating system above.)
-
Clone this repo and place the file in the cloned folder, and unzip (or decompress) the file.
git clone https://github.com/sindyma/udacity-tennis-p3.git
Before running the Jupyter Notebook, ensure all requirements are installed. To do this, navigate to the cloned repo and run the following command. This will install all packages specified in requirements.txt
.
pip install .
Open Navigation.ipynb in the root directory. This Jupyter Notebook trains an RL agent to navigate the environment and collect bananas.
If you aren't able to get started, see if any of the following describe your situation:
- Python version This repo has been developed with python 3.7. Python 2 users may need to switch their kernel and ensure all dependencies have been installed for python 3.7
- Dependencies You may need to install dependencies:
- unity-ml
- numpy
- torch
- matplotlib
Run all the cells in Tennis.ipynb
to train your own agent. At the end of training, your weights will be saved in the folder as <dateandtime>_trained_agent.pth
.
The final section in Tennis.ipynb
runs a trained agent (submission for the project).