Collection of Data Structures

A collection of functions to generate a large variety of structures in high dimensions. These data structures are useful for testing, validating, and improving algorithms used in dimensionality reduction, clustering, machine learning, and visualization.

cardinalR

The cardinalR package provides a collection of functions to generate a large variety of structures in high dimensions.

Installation

You can install the released version of cardinalR from CRAN with:

install.packages("cardinalR") 

The development version from GitHub with:

# install.packages("pak")
pak::pak("JayaniLakshika/cardinalR")

Example

library(cardinalR)

head(mobiusgau, 5)
#> # A tibble: 5 × 4
#>        x1       x2       x3      x4
#>     <dbl>    <dbl>    <dbl>   <dbl>
#> 1  0.517  -0.579   -0.00951 -0.0401
#> 2  0.0142 -0.00841  0.0147   0.0579
#> 3 -0.402   0.506   -0.196   -0.0325
#> 4 -0.218  -0.684    0.0347   0.0497
#> 5  0.225   0.699    0.00575  0.0251

To view the data in high-dimensional space

langevitour(mobiusgau)

Following shows three 2-D projections from the 4-D mobiusgau data.

You can find the high-dimensional view in here.

tSNE (t-distributed Stochastic Neighbor Embedding) and UMAP (Uniform Manifold Approximation and Projection) representations of mobiusgau are shown below. Figures a–c illustrate the t-SNE projections and figures d–f present the corresponding UMAP projections.

About the name

collection of various high-dimensional data structures in R

Copyright

This package is licensed under the MIT license.