MatrixFactorizationBase

class implicit.cpu.matrix_factorization_base.MatrixFactorizationBase(num_threads=0)

MatrixFactorizationBase contains common functionality for recommendation models.

item_factors

Array of latent factors for each item in the training set

user_factors

Array of latent factors for each user in the training set

num_threads

The number of threads to use for batch recommendation calls and fitting the model. Setting to 0 will use all CPU cores on the machine