mattbeth / nba-optimize

Code that show how to optimize a DFS NBA lineup, which is a version of a knapsack problem

nba-optimize

Code that show how to optimize a Daily Fantasy Sports (DFS) NBA lineup, which is a version of a knapsack problem.

Assiciated blog post where I talk about the code.

Three files included, basic_optimize.py, brisk_optimize.py, and fast_optimize.py each show different ways to perform the optimization.

I included csv data files for us to use. I used other code to get the data from nba.com and the info from FanDuel csv files that list the players and their salaries.

I'm assuming you'll have a virtualenv running python3. If so, run the following:

(nbaoptimize) $ pip install -r requirements
  .....
(nbaoptimize) $ python fast_optimize.py 2018-10-24

where the second argument is the data file to use.

Three files are also included to show examples of how the three solutions were done using Jupyter notebooks: example_basic_optimize.ipynb, example_brisk_optimize.ipynb, and example_fast_optimize.ipynb.

To run the Jupyter notebooks, instead of looking at the files here, run

$ jupyter notebook

which opens a new tab in browser where you can click the notebook files and run them yourself.