Clone the project from github:

$ git clone git@github.com:d-rudolf/UdacityML.git &&  cd UdacityML

The branch for project submission is project_submission, not master The fastest way to run the code is to use pipenv. Install pipenv:

$ pip install pipenv

I used python 3.5 throughout the project. Therefore, create a new project with python 3 interpreter:

$ pipenv --python 3.5 && pipenv shell

Install all dependencies from the Pipfile.lock. You should copy the Pipfile.lock file to the pipenv project directory.

$ pipenv install --dev

To run the machine learning code enter

$ python poi_id.py 

For helper functions I wrote a helper.py file, which is in the same folder as the poi_id.py file. I also wrote a small flask app to visualize the data. To run it locally enter

$ cd flask_app && python manage.py runserver 

Then, enter

http://localhost:8080/

in the browser to use the app. The app allows to plot two features on the x- and y-axis. Select two buttons and press the plot button. To clear the plot press the clear button.