[R-Package] mulset - generate dataset intersections in sparse data

[LogIN]

Description

If you have missing values in your data, but you dont want to impute those values or mess with them in any other artificial way,
you can use mulset and generate multiple datasets of those data that you can use latter on in autoML or some other analysis.



Download links

• https://cran.r-project.org/web/packages/mulset/index.html

• https://github.com/LogIN-/mulset

Installation

Code: Select all

install.packages("devtools")
devtools::install_github("LogIN-/mulset", subdir = 'R-package')

Usage Example

Code: Select all

library("mulset")
## Using provided demo data
data(mulsetDemo)
resamples <- mulset(mulsetDemo, exclude = c("outcome", "age", "gender"), 250)

## or from CSV file
## mulsetDemo <-fread("./data/mulsetDemo.csv", header = T, sep = ',', stringsAsFactors = FALSE, data.table = FALSE)
## exclude <- c("come_column")
## resamples <- mulset(mulsetDemo, exclude = exclude, include = c("samples_count", "datapoints"), maxIntersections = 250, hashMethod = "sha1")