1. Learn
    2. /
  2. Courses
    3. /
  3. Building Recommendation Engines in Python

Exercise

Decomposing your matrix

Now that you have prepped your data by centering it and filling in the remaining empty values with 0, you can get around to finding your data's factors. In this exercise, you will break the user_ratings_centered data you generated in the last exercise into 3 factors: U, sigma, and Vt.

  • U is a matrix with a row for each user
  • Vt has a column for each movie
  • sigma is an array of weights that you will need to convert to a diagonal matrix

The user_ratings_centered that you created in the last lesson has been loaded for you.

Instructions 1/2

undefined XP
  • 1
    • Import svds from scipy.sparse.linalg.
    • Decompose user_ratings_pivot_centered into its factor matrices: U, sigma and Vt.
  • 2
    • Convert the sigma array into a diagonal matrix.