This is a general repository for Udacity's MLND program. Each directory represents a different machine learning project. All of these projects rely on the Python data ecosystem & Jupyter notebooks. For best results download the Anaconda Python distribution and create a conda environment.
Follow the instructions at the following link: https://www.continuum.io/downloads
FYI: if you don't want the full Anaconda distribution, download Miniconda. Miniconda does not include all the libraries that the full Anaconda distribution does. Download Miniconda at the following link: http://conda.pydata.org/miniconda.html
These projects were built on top of Python 2.7 (unfortunately)
After Anaconda or Miniconda have been installed it is best to create a virtual environment to interact with any of the projects. This can be done by using the conda environment manager (docs linked below.)
conda create --name mlnd --file requirements.txt
A basic requirements file is included. Note that the rsi project has a separate requirements file. To activate the environment use the following command:
source activate mlnd
conda docs: http://conda.pydata.org/docs/using/envs.html
in order to install Pygame the following command needs to be run:
conda install -c https://conda.binstar.org/quasiben pygame
Here you are adding user quasiben's channel to your Anaconda/Miniconda distribution and installing the package from there. Make certain that you have activated your environment before you install the pygame package.
To learn more about binstar/Anaconda Cloud start at the following link: https://conda.anaconda.org/