Learn R Programming

spray (version 1.0-26)

spray-package: Sparse arrays and multivariate polynomials

Description

Functionality for sparse arrays, with emphasis on their interpretation as multivariate polynomials.

Arguments

Author

Robin K. S. Hankin

Details

Base R has the capability of dealing with arbitrary dimensioned numerical arrays, with the array class.

A sparse array is a type of array in which nonzero elements are stored along with an index vector describing their coordinates. This allows for efficient storage and manipulation as base arrays often require the storing of many zero elements which consume computational and memory resources.

In the package, sparse arrays are represented as objects of class spray. They use the C++ standard template library (STL) map class, with keys being (unsigned) integer vectors, and values floats.

One natural application of sparse arrays, for which the package was written, is multivariate polynomials and the package vignette presents an extended discussion. Note that other interpretations exist: the stokes and weyl packages interpret spray objects as differential forms and elements of a Weyl algebra respectively.

Examples

Run this code

# define a spray using a matrix of indices and a vector of values:
M <- matrix(sample(0:3,21,replace=TRUE),ncol=3)
a <- spray(M,sample(7))

# there are many pre-defined simple sprays:
b <- homog(3,4)

# arithmetic operators work:
a + 2*b
a - a*b^2/4
a+b

#  we can sum over particular dimensions:
asum(a+b,1)

# differentiation is supported:
deriv(a^6,2)

# extraction and replacement work as expected:

b[1,2,1]
b[1,2,1,drop=TRUE]

b[diag(3)] <- 3



Run the code above in your browser using DataLab