Example data structures

library(cardinalR)
library(langevitour)

This illustrates a variety of example data structures that can be generated. These synthetic datasets represent common and challenging shapes found.

Gaussian

A simple spherical Gaussian cluster with very small variance in \(4\text{-}D\) space.

gau_data <- gen_gaussian(n = 500, p = 4, s = diag(4) * 0.01)

langevitour(gau_data, pointSize = 2)

Cube with a hole

Uniform points in a hypercube with a hollow center, useful for detecting voids or gaps.

cube_grd_data <- gen_cubehole(n = 5000, p = 6)

langevitour(cube_grd_data, pointSize = 2)

Circular

A curvilinear loop structure that forms a closed cycle in \(4\text{-}D\).

curvy_cyc_data <- gen_curvycycle(n = 500, p = 4) 

langevitour(curvy_cyc_data, pointSize = 2)

Branching

A tree-like curvilinear structure with six branches in \(6\text{-}D\), ideal for studying bifurcation patterns.

tree_data <- gen_orgcurvybranches(n = 600, p = 6, k = 6) 
#> Warning: The `x` argument of `as_tibble.matrix()` must have unique column names if
#> `.name_repair` is omitted as of tibble 2.0.0.
#> ℹ Using compatibility `.name_repair`.
#> ℹ The deprecated feature was likely used in the cardinalR package.
#>   Please report the issue at
#>   <https://github.com/JayaniLakshika/cardinalR/issues>.
#> This warning is displayed once every 8 hours.
#> Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
#> generated.

langevitour(tree_data, pointSize = 2)

Cone

A pointed cone shape in \(4\text{-}D\), controlled by height and radius ratio.

cone_data <- gen_cone(n = 500, p = 4, h = 5, ratio = 0.5) 

langevitour(cone_data, pointSize = 2)

Conic spiral

A spiral winding around a conical surface in \(5\text{-}D\) space.

spiral_data <- gen_conicspiral(n = 500, p = 5, spins = 2)

langevitour(spiral_data, pointSize = 2)

Grided sphere

A regularly spaced sphere with high point density, useful for manifold learning.

sphere_data <- gen_gridedsphere(n = 500, p = 4)

langevitour(sphere_data, pointSize = 2)

S-curve with a hole

An S-shaped manifold in \(8\text{-}D\) with a missing section, to evaluate resilience to structural gaps.

scurve_data <- gen_scurvehole(n = 600, p = 8) 

langevitour(scurve_data, pointSize = 2)

Pyramid with a hole

A \(4\text{-}D\) pyramid-shaped structure with a void at the center.

pyr_data <- gen_pyrholes(n = 1000, p = 4)

langevitour(pyr_data, pointSize = 2)