sgdgmf.cv.step: Single step of cross-validation for generalized matrix...

In sgdGMF: Estimation of Generalized Matrix Factorization Models via Stochastic Gradient Descent

Single step of cross-validation for generalized matrix factorization models

Description

Internal function running a single step of cross-validation for generalized matrix factorization (GMF) models and calculating some goodness-of-fit measures on the train and test sets.

Usage

sgdgmf.cv.step(
  train,
  test,
  X,
  Z,
  family,
  ncomp,
  maxcomp,
  fold,
  nfolds,
  weights,
  offset,
  method,
  sampling,
  penalty,
  control.init,
  control.alg,
  control.cv
)

Arguments

train	train-set matrix of responses (n \times m)
test	test-set matrix of responses (n \times m)
X	matrix of row fixed effects (n \times p)
Z	matrix of column fixed effects (q \times m)
family	a glm family (see family for more details)
ncomp	ranks of the latent matrix factorization used in cross-validation (default 1 to 10)
maxcomp	maximum rank allowed in the cross-validation exploration
fold	integer number identifying the current fold
nfolds	maximum number of folds in the cross-validation
weights	an optional matrix of weights (n \times m)
offset	an optional matrix of offset values (n \times m), that specify a known component to be included in the linear predictor.
method	estimation method to minimize the negative penalized log-likelihood
sampling	sub-sampling strategy to use if method = "sgd"
penalty	list of penalty parameters (see set.penalty for more details)
control.init	list of control parameters for the initialization (see set.control.init for more details)
control.alg	list of control parameters for the optimization (see set.control.alg for more details)
control.cv	list of control parameters for the cross-validation (see set.control.cv for more details)

Value

Returns a data.frame containing the current number of latent factors in the model (ncomp), the fold identifier (fold), the degrees of freedom, i.e. the number of parameters, of the model (df), the AIC, BIC and deviance (respectively, aic, bic, dev) calculated on the train and test sets.

sgdGMF documentation built on April 3, 2025, 7:37 p.m.