In lab 13, when defining class MatrixFactorizationPredictor(Predictor), we have :
self.movie_features = rng.normal(size=[num_movies, num_features])
self.user_features = rng.normal(size=[num_movies, num_features]) # should be [num_users, num_features]
Isn't the size of self.user_features wrong here ? It won't crash because num_movies > num_users with our dataset and also won't affect the result though.
Lab 13 Solution possible typo
Hi,
In lab 13, when defining class MatrixFactorizationPredictor(Predictor), we have :
self.movie_features = rng.normal(size=[num_movies, num_features])
self.user_features = rng.normal(size=[num_movies, num_features]) # should be [num_users, num_features]
Isn't the size of self.user_features wrong here ? It won't crash because num_movies > num_users with our dataset and also won't affect the result though.
Best regards,
Ali
Hi, yes, good catch. Thanks for letting us know. We’ll fix this in the solution.
1
Add comment