Learn R Programming

RcppAlgos

A collection of high performance functions and iterators implemented in C++ for solving problems in combinatorics and computational mathematics.

Featured Functions

  • {combo|permute}General: Generate all combinations/permutations of a vector (including multisets) meeting specific criteria.
  • {partitions|compositions}General: Efficient algorithms for partitioning numbers under various constraints
  • {expandGrid|comboGrid}: Generate traditional Cartesian product as well as the product where order does not matter.
  • {combo|permute|partitions|compositions|expandGrid|comboGroups}Sample: Generate reproducible random samples
  • {combo|permute|partitions|compositions|expandGrid|comboGroups}Iter: Flexible iterators allow for bidirectional iteration as well as random access.
  • primeSieve: Fast prime number generator
  • primeCount: Prime counting function using Legendre's formula

The primeSieve function and the primeCount function are both based off of the excellent work by Kim Walisch. The respective repos can be found here: kimwalisch/primesieve; kimwalisch/primecount

Additionally, many of the sieving functions make use of the fast integer division library libdivide by ridiculousfish.

Benchmarks

Installation

install.packages("RcppAlgos")

## install the development version
devtools::install_github("jwood000/RcppAlgos")

Usage

Combinatorics Basics

## Find all 3-tuples combinations of 1:4
comboGeneral(4, 3)
#>      [,1] [,2] [,3]
#> [1,]   1    2    3
#> [2,]   1    2    4
#> [3,]   1    3    4
#> [4,]   2    3    4


## Alternatively, iterate over combinations
a = comboIter(4, 3)
a@nextIter()
#> [1] 1 2 3

a@back()
#> [1] 2 3 4

a[[2]]
#> [1] 1 2 4


## Pass any atomic type vector
permuteGeneral(letters, 3, upper = 4)
#>      [,1] [,2] [,3]
#> [1,] "a"  "b"  "c"
#> [2,] "a"  "b"  "d"
#> [3,] "a"  "b"  "e"
#> [4,] "a"  "b"  "f"


## Generate a reproducible sample
comboSample(10, 8, TRUE, n = 5, seed = 84)
#>      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
#> [1,]    3    3    3    6    6   10   10   10
#> [2,]    1    3    3    4    4    7    9   10
#> [3,]    3    7    7    7    9   10   10   10
#> [4,]    3    3    3    9   10   10   10   10
#> [5,]    1    2    2    3    3    4    4    7

Integer Partitions and Constraints

## Flexible partitioning algorithms
partitionsGeneral(0:5, 3, freqs = rep(1:2, 3), target = 6)
#>      [,1] [,2] [,3]
#> [1,]    0    1    5
#> [2,]    0    2    4
#> [3,]    0    3    3
#> [4,]    1    1    4
#> [5,]    1    2    3


## And compositions
compositionsGeneral(0:3, repetition = TRUE)
#>      [,1] [,2] [,3]
#> [1,]    0    0    3
#> [2,]    0    1    2
#> [3,]    0    2    1
#> [4,]    1    1    1


## Get combinations such that the product is between
## 3600 and 4000 (including 3600 but not 4000)
comboGeneral(5, 7, TRUE, constraintFun = "prod",
             comparisonFun = c(">=","<"),
             limitConstraints = c(3600, 4000),
             keepResults = TRUE)
#>      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
#> [1,]    1    2    3    5    5    5    5 3750
#> [2,]    1    3    3    4    4    5    5 3600
#> [3,]    1    3    4    4    4    4    5 3840
#> [4,]    2    2    3    3    4    5    5 3600
#> [5,]    2    2    3    4    4    4    5 3840
#> [6,]    3    3    3    3    3    3    5 3645
#> [7,]    3    3    3    3    3    4    4 3888


## We can even iterate over constrained cases. These are
## great when we don't know how many results there are upfront.
## Save on memory and still at the speed of C++!!
p = permuteIter(5, 7, TRUE, constraintFun = "prod",
                comparisonFun = c(">=","<"),
                limitConstraints = c(3600, 4000),
                keepResults = TRUE)

## Get the next n results
t = p@nextNIter(1048)

## N.B. keepResults = TRUE adds the 8th column
tail(t)
#>         [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
#> [1043,]    5    4    4    3    4    1    4 3840
#> [1044,]    5    4    4    3    4    4    1 3840
#> [1045,]    5    4    4    4    1    3    4 3840
#> [1046,]    5    4    4    4    1    4    3 3840
#> [1047,]    5    4    4    4    3    1    4 3840
#> [1048,]    5    4    4    4    3    4    1 3840

## Continue iterating from where we left off
p@nextIter()
#> [1]    5    4    4    4    4    1    3 3840

p@nextIter()
#> [1]    5    4    4    4    4    3    1 3840

p@nextIter()
#> [1]    2    2    3    3    4    5    5 3600

## N.B. totalResults and totalRemaining are NA because there is no
## closed form solution for determining this.
p@summary()
#> $description
#> [1] "Permutations with repetition of 5 choose 7 where the prod is between 3600 and 4000"
#> 
#> $currentIndex
#> [1] 1051
#> 
#> $totalResults
#> [1] NA
#> 
#> $totalRemaining
#> [1] NA

Cartesian Products

## Base R expand.grid returns a data.frame by default
## and varies the first column the fastest
bR = expand.grid(rep(list(1:3), 3))
head(bR, n = 3)
#>   Var1 Var2 Var3
#> 1    1    1    1
#> 2    2    1    1
#> 3    3    1    1

tail(bR, n = 3)
#>    Var1 Var2 Var3
#> 25    1    3    3
#> 26    2    3    3
#> 27    3    3    3


## RcppAlgos::expandGrid returns a matrix if the input is of
## the same class, otherwise it returns a data.frame. Also
## varies the first column the slowest.
algos = expandGrid(rep(list(1:3), 3))
head(algos, n = 3)
#>      Var1 Var2 Var3
#> [1,]    1    1    1
#> [2,]    1    1    2
#> [3,]    1    1    3

tail(algos, n = 3)
#>       Var1 Var2 Var3
#> [25,]    3    3    1
#> [26,]    3    3    2
#> [27,]    3    3    3


## N.B. Since we are passing more than one type, a data.frame is returned
expandGrid(
    c(rep(list(letters[1:3]), 3), list(1:3)),
    upper = 3
)
#>   Var1 Var2 Var3 Var4
#> 1    a    a    a    1
#> 2    a    a    a    2
#> 3    a    a    a    3


## With RcppAlgos::comboGrid order doesn't matter, so c(1, 1, 2),
## c(1, 2, 1), and c(2, 1, 1) are the same.
comboGrid(rep(list(1:3), 3))
#>       Var1 Var2 Var3
#>  [1,]    1    1    1
#>  [2,]    1    1    2
#>  [3,]    1    1    3
#>  [4,]    1    2    2
#>  [5,]    1    2    3
#>  [6,]    1    3    3
#>  [7,]    2    2    2
#>  [8,]    2    2    3
#>  [9,]    2    3    3
#> [10,]    3    3    3


## If you don't want any repeats, set repetition = FALSE
comboGrid(rep(list(1:3), 3), repetition = FALSE)
#>      Var1 Var2 Var3
#> [1,]    1    2    3

Partitions of Groups

Efficiently partition a vector into groups with comboGroups. For example, the code below finds all possible pairings of groups of size 3 vs groups of size 2 (See this stackoverflow post: Find all possible team pairing schemes).

players <- c("Ross", "Bobby", "Max", "Casper", "Jake")
comboGroups(players, grpSizes = c(2, 3))
#>       Grp1     Grp1     Grp2    Grp2     Grp2    
#>  [1,] "Ross"   "Bobby"  "Max"   "Casper" "Jake"  
#>  [2,] "Ross"   "Max"    "Bobby" "Casper" "Jake"  
#>  [3,] "Ross"   "Casper" "Bobby" "Max"    "Jake"  
#>  [4,] "Ross"   "Jake"   "Bobby" "Max"    "Casper"
#>  [5,] "Bobby"  "Max"    "Ross"  "Casper" "Jake"  
#>  [6,] "Bobby"  "Casper" "Ross"  "Max"    "Jake"  
#>  [7,] "Bobby"  "Jake"   "Ross"  "Max"    "Casper"
#>  [8,] "Max"    "Casper" "Ross"  "Bobby"  "Jake"  
#>  [9,] "Max"    "Jake"   "Ross"  "Bobby"  "Casper"
#> [10,] "Casper" "Jake"   "Ross"  "Bobby"  "Max"

Computational Mathematics

## Generate prime numbers
primeSieve(50)
#> [1]  2  3  5  7 11 13 17 19 23 29 31 37 41 43 47

## Many of the functions can produce results in
## parallel for even greater performance
p = primeSieve(1e15, 1e15 + 1e8, nThreads = 4)

head(p)
#> [1] 1000000000000037 1000000000000091 1000000000000159
#> [4] 1000000000000187 1000000000000223 1000000000000241
tail(p)
#> [1] 1000000099999847 1000000099999867 1000000099999907
#> [4] 1000000099999919 1000000099999931 1000000099999963


## Count prime numbers less than n
primeCount(1e10)
#> [1] 455052511

## Get the prime factorization
set.seed(24028)
primeFactorize(sample(1e15, 3), namedList = TRUE)
#> $`701030825091514`
#> [1]             2           149 2352452433193
#> 
#> $`83054168594779`
#> [1]  3098071 26808349
#> 
#> $`397803024735610`
#> [1]            2            5           13           13 235386405169

Further Reading

Why RcppAlgos but no Rcpp?

Previous versions of RcppAlgos relied on Rcpp to ease the burden of exposing C++ to R. While the current version of RcppAlgos does not utilize Rcpp, it would not be possible without the myriad of excellent contributions to Rcpp.

Contact

If you would like to report a bug, have a question, or have suggestions for possible improvements, please file an issue.

Copy Link

Version

Install

install.packages('RcppAlgos')

Monthly Downloads

1,528

Version

2.9.3

License

GPL (>= 2)

Issues

Pull Requests

Stars

Forks

Maintainer

Joseph Wood

Last Published

February 3rd, 2025

Functions in RcppAlgos (2.9.3)

Constraints-class

S4-class for Exposing C++ Constraints Class
comboIter

Combination and Permutation Iterator
comboCount

Number of combinations/permutations
comboRank

Rank Combinations and Permutations
comboGroups

Partition a Vector into Groups
numDivisorSieve

Apply Divisor Function to Every Element in a Range
Partitions-class

S4-class for Exposing C++ Partitions Class
comboGeneral

Generate Combinations and Permutations of a Vector with/without Constraints
comboGroupsCount

Number of Partitions of a Vector into Groups
partitionsCount

Number of Partitions/Compositions
RcppAlgos-package

High Performance Tools for Combinatorics and Computational Mathematics
partitionsRank

Rank Partitions/Compositions
Cartesian-class

S4-class for Exposing C++ Cartesian Class
comboSample

Sample Combinations and Permutations
comboGroupsIter

comboGroups Iterator
Combo-class

S4-classes for Exposing C++ Combinatorial Classes
comboGroupsSample

Sample Partitions of a Vector into Groups
eulerPhiSieve

Apply Euler's Phi Function to Every Element in a Range
comboGrid

Unordered Cartesian Product
stdThreadMax

Max Number of Concurrent Threads
expandGrid

Cartesian Product
divisorsRcpp

Vectorized Factorization (Complete)
partitionsSample

Sample Partitions/Compositions
expandGridCount

Count of the Cartesian Product
expandGridSample

Sample the Cartesian Product
isPrimeRcpp

Vectorized Primality Test
expandGridIter

expandGrid Iterator
divisorsSieve

Generate Complete Factorization for Numbers in a Range
partitionsGeneral

Generate Partitions/Compositions
primeCount

Prime Counting Function \(\pi(x)\)
primeFactorize

Vectorized Prime Factorization
partitionsIter

Partition/Composition Iterator
primeFactorizeSieve

Generate Prime Factorization for Numbers in a Range
primeSieve

Generate Prime Numbers
ComboGroups-class

S4-class for Exposing C++ ComboGroups Class