Matrix Factorization for Collaborative Filtering: Linear to Non-Linear Models in Python

Accompanying repo for this article. Code is found in matrix_factorization.ipynb and ratings.csv contains the ratings for the MovieLens 100K dataset.

Running the Code

Make sure to have at least Python3.8 (might be fine with an older version of Python3). Dependencies listed in requirements.txt.

Run through matrix_factorization.ipynb; many comments included to help. Follow the article linked above to understand it in better detail.

Example MovieLens Performance

Other Article Images