In ryan-heslin/matador: Linear Algebra Visualization

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(matador)
ggplot2::theme_set(ggplot2::theme_minimal())

Linear algebra is a formidably abstract subject, one that conceals its beauty behind inscrutable matrices.

Matador can help. Born of procrastination on my linear algebra homework, this modest package provides a suite of handy functions for working with and plotting matrices. Other matrix algebra packages specialize in efficient computation, but matador instead focuses on convenience and visualization.

Using the mat2latex function, you can easily translate matrix computations in R into LaTeX code suitable for Rmarkdown:

A <- matrix(-1:2, nrow = 2)
B <- matrix(rep(0.5, 4), nrow = 2)
C <- diag(x = 4, nrow = 2)
ABC <- matador::compose_trans(list(C, B, A))
mats <- lapply(list(C, B, A, ABC), matador::mat2latex, sink = TRUE)
invisible(sapply(mats, function(x) cat(x, sep = "\n")))

With the mat_pows function, you can multiply a matrix by itself as many times as you like:

m <- matrix(c(cos(pi / 3), sin(pi / 3), -sin(pi / 3), cos(pi / 3)), nrow =
              2)
lapply(mat_pows(m, 1:7), round, digits = 3) 

You can solve a system of equations visually using the plot_lines function

set.seed(1)

m2 <- matrix(sample(-10:10, 10), nrow = 5)
plot_lines(m = m2, b = sample(0:10, 5)) 

There are also convenience functions to simplify common computations. square, for instance, will automatically construct a square matrix:

square(letters[1:9])

But the centerpiece is plot_transform, a flexible function that plots a set of vectors before and after a linear transformation:

plot_transform(trans = list(diag(x=3, nrow =2), 
                            matrix(c(-1, 0, 0, 1), nrow =2), matrix(c(1, 0, 2, 1), nrow = 2))) 

Other functions can help you do things like determine whether matrices compute, plot the orthogonal decomposition of a vector, and more.

I hope matador will help you spend less time poring over matrices and more learning the fascinating nuts and bolts of linear algebra.

ryan-heslin/matador documentation built on Dec. 22, 2021, 8:17 p.m.