The tiler
R-package provides functions for constrained randomisation
of matrices using tiles. The package also includes an interactive
software for data exploration.
To install this R-package, proceed as follows.
First install the devtools
-package and load it in R:
install.packages("devtools")
library(devtools)
Then install the tiler
package
install_github("aheneliu/tiler")
After installation, start R and load the package using
library(tiler)
Here is a brief example of how to use the R-package.
library(tiler)
## Create a data matrix with a special structure
nr <- 11
nc <- 13
x <- apply(matrix(rep(seq.int(nc), nr), nrow = nc, ncol = nr), 1, function(i) (i-1)*10 ) + matrix(rep(seq.int(nr), nc), nr, nc)
## Initialise a tiling where each column is a separate tile
tiling <- init_tiling(dim(x), columnwise_tiles = TRUE)
## Create some tiles
t1 <- create_tile(rowset = c(1, 2, 3, 4), colset = c(5, 6, 7), dims = dim(x))
t2 <- create_tile(rowset = c(4, 5, 6, 7), colset = c(1, 2, 3, 8, 9), dims = dim(x))
t3 <- create_tile(rowset = c(3, 4, 5), colset = c(2, 3, 4, 5, 6), dims = dim(x))
t4 <- create_tile(rowset = c(2, 3, 4, 7, 8, 9), colset = c(2, 3, 4, 8, 9, 10), dims = dim(x))
t5 <- create_tile(rowset = c(5, 6, 7, 8, 9, 11), colset = c(11, 12, 13), dims = dim(x))
## Add the tiles to the tiling
tiling <- add_tile(tiling, t1)
tiling <- add_tile(tiling, t2)
tiling <- add_tile(tiling, t3)
tiling <- add_tile(tiling, t4)
tiling <- add_tile(tiling, t5)
## View the tiling
print_tiling(tiling)
## Obtain a permuted version of the data, constrained by the tiling
x2 <- permute_data(x, tiling)
The package also includes an interactive application based on shiny. To start the application, proceed as follows in R.
library(tiler)
tiler::run_tiler()
This should also automatically launch a web browser with the application.
Otherwise you can open a web browser and point it to http://127.0.0.1:7000/
.
The tiler
R-package is licensed under the MIT-license.