===============================

A Python module for parsing and analyzing sports statistics.

kind of optional

1. create and activate a virtual environment

   • for vanilla Python:

   $ pip install virtualenv $ virtualenv sportstat $ source sportstat/bin/activate

   • with Anaconda:

   $ conda create -n sportstat anaconda $ source activate sportstat

Note: when using Anaconda, you may choose to limit the packages to just those indicated in the requirements.txt file by specifying:

$ conda create -n sportstat --file requirements.txt

For developers; to update requirements.txt after installing packages (using pip or conda install), run:

$ conda list -e > requirements.txt

to overwrite the requirements file.

2. Install dependencies

   $ pip install -r requirements.txt

3. Install sportstat python package

   • first build the package:

   $ python setup.py build

   • then, if only deploying (not developing):

   $ python setup.py install

   • if developing or testing:

   $ python setup.py develop

The module sportstat contains useful query and analysis functions along with Mongo document models. The module sportstat_server contains a Flask application for serving the data REST API.

TODO: There is an interdependency--because MongoEngine uses Flask for its configuration, and we want the same MongoEngine wrapper when working on analytics and on the REST server, we end up pulling the MongoEngine db instance from the sportstat_server module into the models module when defining the Mongo document models. This should not be a problem as long as these two projects are together.

To analyze/parse data using jupyter notebooks, make sure your jupyter instance is running in the same environment that you installed sportstat and sportstat_server.