componentCVPatterns: Cross-Validation of Simplivariate Component Patterns

In SIMPLICA: Biclustering via Simplivariate Component Analysis

Cross-Validation of Simplivariate Component Patterns

Description

Performs pure cross-validation over specified patterns with mandatory fitters. This function evaluates different pattern fitting models using cross-validation to determine the best model for a given data subset.

Usage

componentCVPatterns(
  df,
  rows,
  cols,
  patternFunctions,
  patternFitters,
  preferenceOrder = names(patternFunctions),
  nRepeats = 40,
  testFraction = 0.2,
  minCellsForModels = 25,
  parsimonyMargin = 0.05,
  requireFitters = TRUE,
  verbose = FALSE
)

Arguments

df	A matrix or data frame containing the data
rows	Row indices to subset from df
cols	Column indices to subset from df
patternFunctions	A named list of pattern functions to evaluate
patternFitters	A named list of fitter functions corresponding to each pattern
preferenceOrder	Character vector specifying the preference order of patterns (default: names of patternFunctions)
nRepeats	Integer, number of cross-validation repeats (default: 40)
testFraction	Numeric, fraction of data to use for testing in each CV fold (default: 0.2)
minCellsForModels	Integer, minimum number of cells required for reliable CV (default: 25)
parsimonyMargin	Numeric, margin for parsimony selection as fraction (default: 0.05)
requireFitters	Logical, whether to require fitters for all patterns (default: TRUE)
verbose	Logical, whether to print progress messages (default: FALSE)

Value

A list containing:

decision	Character, the selected best pattern name
reason	Character, explanation of the selection reasoning
cv	Data frame with CV summary statistics for each model
repeats	Data frame with detailed results from each CV repeat
meta	List with metadata about the CV procedure

SIMPLICA documentation built on Sept. 11, 2025, 1:08 a.m.