big_read: Read big data or a big.matrix object

In pchc: Bayesian Network Learning with the PCHC and Related Algorithms

Read big data or a big.matrix object

Description

Read big data or a big.matrix object.

Usage

big_read(big_path, select, header = TRUE, sep = ",")

Arguments

big_path	The path (including the name) where the big.matrix object is.
select	Indices of columns to read (sorted). The length of select will be the number of columns of the resulting filebacked Big Matrix.
header	If there are column names, then this should be TRUE.
sep	A field delimiter, for example ";" or "," (comma separated). See also read.csv for more information.

Details

The data (matrix) which will be read and compressed into a big.matrix object must be of type "numeric". I tested it and it works with "integer" as well. But, in general, bear in mind that only matrices will be read. I have not tested with data.frame for example. However, in the help page of "bigmemory" this is mentioned: Any non-numeric entry will be ignored and replaced with NA, so reading something that traditionally would be a data.frame won't cause an error. A warning is issued. In all cases, the big.matrix is turned into a Filebacked Big Matrix (FBM) of type 'double' the object size is alwasy 680 bytes! If the initial dataset has row names these will be ignored and a column with NAs will apear. So check your final FBM matrix. For more information see the "bigmemory" and "bigstatsr" packages.

Value

A Filebacked Big Matrix (FBM) matrix.

Author(s)

Michail Tsagris.

R implementation and documentation: Michail Tsagris [e-mail].

See Also

big_cor, fedhc.skel, mmhc.skel

Examples

x <- matrix( runif(100 * 5, 1, 100), ncol = 5 )

pchc documentation built on April 4, 2025, 1:11 a.m.