RandomFArray-class: DelayedArray of random F-distributed values

In LTLA/DelayedRandomArray: Delayed Arrays of Random Values

Description

A DelayedArray subclass that performs on-the-fly sampling of F-distributed values.

Usage

RandomFArraySeed(dim, df1, df2, ncp, chunkdim = NULL)

## S4 method for signature 'RandomFArraySeed'
DelayedArray(seed)

RandomFArray(dim, df1, df2, ncp, chunkdim = NULL)

Arguments

dim	Integer vector of positive length, specifying the dimensions of the array.
df1, df2, ncp	Numeric vector used as the argument of the same name in qf. Alternatively, a numeric array-like object with the same dimensions as dim. If ncp is missing, a central F distribution is assumed.
chunkdim	Integer vector of length equal to dim, containing the dimensions of each chunk.
seed	A RandomFArraySeed object.

Value

All constructors return an instance of a RandomFArray object, containing random draws from a exponential distribution with the specified parameters.

Author(s)

Aaron Lun

See Also

The RandomArraySeed class, for details on chunking and the distributional parameters.

Examples

X <- RandomFArraySeed(c(1e5, 1e5), df1=1, df2=10)
Y <- DelayedArray(X)
Y

# Fiddling with the distribution parameters:
X2 <- RandomFArraySeed(c(1e5, 1e5), df1=runif(1e5)*10, df2=10)
Y2 <- DelayedArray(X2)
Y2

# Using another array as input:
library(Matrix)
ncp <- rsparsematrix(1e5, 1e5, density=0.00001)
ncp <- abs(DelayedArray(ncp)) + 1
X3 <- RandomFArraySeed(c(1e5, 1e5), df1=1, df2=10, ncp=ncp)
Y3 <- DelayedArray(X3)
Y3

LTLA/DelayedRandomArray documentation built on Dec. 18, 2021, 3:40 a.m.