Matrix Factorization Rating Prediction
~10 mincode completion
Implement predict_ratings(mu, user_bias, item_bias, P, Q, users, items) returning one prediction per (users[j], items[j]) pair.
Where this shows up
~10 min
••••••••••••••••
8 employers weight this skill
4 quant funds, 2 defense companies, 2 autonomy companies. Top match scores 93.
Python
import numpy as np
def predict_ratings(mu, user_bias, item_bias, P, Q, users, items):
"""
Biased matrix-factorization predictions for a list of user-item pairs.
Args:
mu: global mean rating
user_bias: (U,) bias of each user
item_bias: (I,) bias of each item
P: (U, k) user embeddings
Q: (I, k) item embeddings
users: (m,) user id of each pair
items: (m,) item id of each pair
Returns:
(m,) array of predicted ratings
"""
# YOUR CODE HERE
pass