bdPCA_hdf5: PCA Descomposition

In BigDataStatMeth: Statistical Methods and Algorithms for Big Data

PCA Descomposition

Description

Compute PCA

Usage

bdPCA_hdf5(
  filename,
  group,
  dataset,
  ncomponents = 0L,
  bcenter = FALSE,
  bscale = FALSE,
  k = 2L,
  q = 1L,
  rankthreshold = 0,
  force = FALSE,
  threads = NULL
)

Arguments

filename	string, file name where dataset is stored
group	string group name where dataset is stored in file
dataset	string dataset name with data to perform PCA
ncomponents	integer, number of components to be computed, by default ncomponents = 0, all components are computed
bcenter	logical value if true data is centered to zero
bscale	logical value, if true data is scaled
k	number of local SVDs to concatenate at each level, performance parameter
q	number of levels to compute SVD for PCA, performance parameter
rankthreshold	double, threshold used to determine the range of the array. The matrix rank is equal to the number of singular values different from the threshold. By default, threshold = 0 is used to get the matrix rank , but it can be changed to an approximation of 0.
force	logical value, if true, the SVD is forced to be computed although the SVD exists
threads	integer number of threads used to run PCA

Value

original file with results in folder PCA/<datasetname>

BigDataStatMeth documentation built on March 30, 2022, 1:07 a.m.