Lunar Lander

This project focuses on using reinforcement learning to achieve highest score in LunarLander-v2.

Setup Instructions:

1. Make sure you have latest version of virtual env, pip install --upgrade virtualenv
2. Make sure python3 is installed on your machine. To check, type python3. If not available, setup from here.
3. Create a virtual env by virtualenv -p python3 venv at the root of this repo (can be anywhere else too, but make sure not to commit this folder). This will create a directory venv, containing default dependencies for python3. If you create this folder at the root of this repo, it will be ignored by git before pushing.
4. Activate virtual env by source venv/bin/activate. This will activate the virtual env. Running python3 will open console with default dependencies from python.
5. Install dependencies for this project by pip3 install -r requirements.txt --no-cache. This will install all the dependencies inside the python virtual environment.
6. Note: Pip will have installed matpltlib, there's an issue with matpltlib with mac os. Follow the steps here to fix this issue.

You are ready! We recommend using pycharm community edition. Once installed, follow the following steps to point project interpreter for pycharm:

1. Go to Preferences, by shortcut Cmd + , or in Menu Bar PyCharm > Preferences...
2. Go to Project Interpreter under Project:lunar-lander
3. Click on settings cog wheel on right side of top Project Interpreter path specification and click on Add.... Choose existing environment and add the path of the one you created. Here's detailed description of how to configure venv in pycharm.

Google Cloud:

1. Prior to any step, please make sure you have dependencies specific to debian machine for installing gym.

apt-get install libjpeg-dev cmake swig python-pyglet python3-opengl libboost-all-dev
libsdl2-2.0.0 libsdl2-dev libglu1-mesa libglu1-mesa-dev libgles2-mesa-dev
freeglut3 xvfb libav-tools

2. Use requirements-google-cloud.txt to install requirements specific to debian.

Contributing:

Please follow the following guidelines to contribute.

1. Create a new branch from master, make changes and open a PR.
2. Get it reviewed by contributers and we merge it in collaboration.

Contributors:

Abhishek Bharani, Amey Naik and Prabhjot Singh Rai.