In krzjoa/matricks: Useful Tricks for Matrix Manipulation

knitr::opts_chunk$set(echo = TRUE)

matricks package in 0.8.2 version has been released on CRAN! In this post I will present you, what are advantagese of using matricks and how you can use it.

Creating matrices

The main function the package started with is m. It's a smart shortcut for creating matrices, especially usefull if you want to define a matrix by enumerating all the elements row-by-row. Typically, if you want to create a matrix in R, you can do it using base function called matrix.

matrix(c(3 ,4, 7,
         5, 8, 0,
         9, 2, 1), nrow = 3, byrow = TRUE)

Although it's a very simple opeartion, the funtion call doesn't look tidy. Alternaively, we can use tibble with its frame_matrix function, defining column names with formulae first.

library(tibble)
frame_matrix(~ c1, ~ c2, ~ c3,
                3,    4,    7,
                5,    8,    0,
                9,    2,    1)

However, it's still not a such powerfull tool as matricks::m function is. Let's see an example.

library(matricks)
m(3 ,4, 7|
  5, 8, 0|
  9, 2, 1)

As simple as that! We join following rows using | operator. m function is very flexible and offers you much more than before mentioned ones.

m(1:3 | 4, 6, 7 | 2, 1, 4)

And here and example with bindig multiple matrices together:

mat1   <- diag(1, 3, 3)
mat2  <- antidiag(1, 3, 3) * 3
m(mat1, mat2|
  mat2, mat1)

By the way, antidiag function can be found in the matricks package too.

Setting & accessing values

These code

mat <- matrix(0, 3, 3)
mat[1, 2] <- 0.3
mat[2, 3] <- 7
mat[3, 1] <- 13
mat[2, 2] <- 0.5
mat

can be replaced with:

mat <- matrix(0, 3, 3)
mat <- set_values(mat,
                  c(1, 2) ~ 0.3,
                  c(2, 3) ~ 7,
                  c(3, 1) ~ 13,
                  c(2, 2) ~ 0.5)
mat

In some cases, traditional way we access a matrix element in R may be inconvenient. Consider situation shown below:

sample.matrix <- matrix(1, 3, 3)
matrix.indices <- list(c(1, 1), c(1, 2), 
                       c(1, 3), c(2, 2),
                       c(3, 1), c(3, 3))

for (idx in matrix.indices) {
  sample.matrix[idx[1], idx[2]] <- sample.matrix[idx[1], idx[2]] + 2
}

sample.matrix

It can be expressed conciser using matrix at function.

sample.matrix <- matrix(1, 3, 3)
matrix.indices <- list(c(1, 1), c(1, 2), 
                       c(1, 3), c(2, 2),
                       c(3, 1), c(3, 3))

for (idx in matrix.indices) {
  at(sample.matrix, idx) <- at(sample.matrix, idx) + 2
}
sample.matrix

Plotting matrix

matrix objects haven't had good automatized plotting function until now. Let's create and plot a sample matrix of random values.

rmat <- runifm(3, 3)
print(rmat)
plot(rmat)

And here the same using a matrix of random boolean values (rboolm).

set.seed(7)
rmat <- rboolm(3, 3)
print(rmat)
plot(rmat)

Operators

matricks contains a family of operators, which allows you to perform column-/row-wise operation (addition/subtraction/multiplication/division) between matrix and vector.

mat <- m(1, 2, 3|
         4, 5, 6|
         7, 8, 9)
mat
vec <- v(1:3)
vec

If we try to do a column-wise multiplication, we ecounter a problem.

mat * vec

We can bypass this error using %m% function. It does what we want!

mat %m% vec

There are also several other operators available.

mat %d% vec
mat %+% vec
mat %-% vec

I encourage you to familiarize with matricks. Visit matrix documentation site and learn more!

krzjoa/matricks documentation built on March 5, 2020, 8:38 p.m.