train_test_split

sparse user-item interation matrix. Internally <code>Matrix::TsparseMatrix</code> is used.

- proportion of the observations for each user to keep as "test" data.

test_proportion

Basic splitting of the user-item interaction matrix into train and testing part.
Useful for when data doesn't have time dimension.
Usually during model tuning it worth to keep some <code>x</code> matrix as hold-out data set.
Then this <code>x</code> could be splitted in 2 parts - train and test.
Model tries to predict test data using train

Implements many (sparse) matrix factorization algorithms.
Focus is on applications for recommender systems.
List of algorithms:
1) Weighted Regularazied Matrix Factorization with Alternating Least Squares (ALS)
for implicit feedback (inculding approximate Conjugate Gradient solver).
Optional non-negativity (NNMF, non-negative matrix factorization).
2) Regularazied Matrix Factorization with ALS for explicit feedback
Optional non-negativity (NNMF, non-negative matrix factorization).
3) Soft-SVD via fast ALS
4) Soft-Impute via fast ALS and solution in SVD form
5) LinearFlow method which learns item-item similarity matrix from the data
All algorithms work on sparse matrices. Extensively use BLAS and LAPACK
and parallelized with OpenMP where applicable.
Implementations are reasonably fast and nicely work with large
datasets (millions of rows and millions of columns).

Last chance! 50% off unlimited learning

train_test_split: Creates cross-validation set from user-item interactions

Description

Usage

Arguments