mf.rst

Classic Matrix Factorization

LKPY provides classical matrix factorization implementations.

Common Support

lenskit.algorithms.mf_common

The :pymf_common module contains common support code for matrix factorization algorithms. These classes, :pyMFPredictor and :pyBiasMFPredictor, define the parameters that are estimated during the :py.Algorithm.fit process on common matrix factorization algorithms.

MFPredictor

BiasMFPredictor

Alternating Least Squares

lenskit.algorithms.als

LensKit provides alternating least squares implementations of matrix factorization suitable for explicit feedback data. These implementations are parallelized with Numba, and perform best with the MKL from Conda.

lenskit.algorithms.als

BiasedMF

ImplicitMF

FunkSVD

lenskit.algorithms.funksvd

FunkSVD is an SVD-like matrix factorization that uses stochastic gradient descent, configured much like coordinate descent, to train the user-feature and item-feature matrices.

FunkSVD