Package {sillysplines}

Add noise variables to create high-dimensional data

Description

This function generates samples from a uniform distribution to be used as noise variables. This can be used to assess the effect of high-dimensions on a model fit, and explainers.

Usage

add_noise_vars(data, n_vars = 2)

Arguments

Value

A data frame with three+n_vars columns, like:

x1

Random uniform coordinate

x2

Random uniform coordinate

x3

From the original data x

x4

From the original data y

class

Binary class label ("Above" or "Below")

Examples

coords <- matrix(c(0.2, 0.3,
                   0.4, 0.5,
                   0.6, 0.7), ncol = 2, byrow = TRUE)

df <- create_data(coord = coords, n_samples = 500, seed = 717)

df <- add_noise_vars(df, 2)
head(df)

Classify Points Relative to a Boundary Line

Description

Classifies data points as being above or below a boundary defined by a set of coordinates. The function uses piecewise linear interpolation to determine the boundary's y-value at each x-coordinate in the data, then classifies points based on whether they lie above or below this boundary.

Usage

classify_boundary(data, boundary_coord, classes = c("Above", "Below"))

Arguments

Details

The function uses stats::approxfun() with rule = 2 for piecewise linear interpolation. This means:

• Between boundary points, linear interpolation is used

• For x-values outside the range of boundary coordinates, the boundary is extrapolated linearly using the slope at the nearest endpoint

Points exactly on the boundary (y == boundary_y) are classified as "Below" (the second class).

Value

A character vector of the same length as the number of rows in data, containing class labels for each point.

See Also

approxfun for details on interpolation

Examples

# Create sample data
data <- data.frame(
  x = c(1, 2, 3, 4, 5),
  y = c(2, 4, 3, 5, 1)
)

# Define a simple linear boundary
boundary <- data.frame(
  x = c(1, 5),
  y = c(2, 4)
)

# Classify points
classify_boundary(data, boundary)

# Use custom class labels
classify_boundary(data, boundary, classes = c("High", "Low"))

# Complex boundary with multiple segments
boundary_complex <- data.frame(
  x = c(1, 2, 3, 4, 5),
  y = c(1, 3, 2, 4, 3)
)
classify_boundary(data, boundary_complex)

Create a 2D Synthetic Dataset With Class Labels

Description

Generates a synthetic two-dimensional dataset using random uniform values. Each point is assigned a class label based on whether it lies above all coordinate thresholds provided. Coordinates can be supplied directly as a matrix/data frame or indirectly via a JSON file containing an array of coordinate pairs.

Usage

create_data(
  coord_filepath = NULL,
  coord = NULL,
  n_samples = 10000,
  type = "uniform",
  seed = 1835,
  gap = 0,
  r = 0
)

Arguments

Details

A point is labelled class = 1 if it is above all coordinate thresholds such that:

x > coord[,1] \; \text{and} \; y > coord[,2]

Otherwise, the point is labelled 0.

Value

A data frame with three columns:

x

Random uniform x-coordinate

y

Random uniform y-coordinate

class

Binary class label ("Above" or "Below")

Examples

coords <- matrix(c(0.2, 0.3,
                   0.4, 0.5,
                   0.6, 0.7), ncol = 2, byrow = TRUE)

df <- create_data(coord = coords, n_samples = 500, seed = 717)
head(df)

Create a box to draw silly spline

Description

This creates a simple htmlwidget to be used

Usage

sillysplines(width = 640, height = 640, elementId = "app")

Arguments

Value

Returns an htmlWidget to be used within an HTML document

Examples

 sillysplines()

Shiny bindings for sillysplines

Description

Output and render functions for using sillysplines within Shiny applications and interactive Rmd documents.

Usage

sillysplinesOutput(outputId, width = "100%", height = "400px")

renderSillysplines(expr, env = parent.frame(), quoted = FALSE)

Arguments

Value

Returns an htmlwidgets::shinyWidgetOutput to be used within the UI of a shiny app.

Examples

library(shiny)
app <- shinyApp(
  ui = fluidPage(sillysplinesOutput('splines')),
  server = function(input, output) {
    splines_widget <- sillysplines()
    output$splines= renderSillysplines(splines_widget)
  }
)

if (interactive()) app

Trim data to be have correlated predictors

Description

This function removes points outside of a elliptical region

Usage

trim_shape(data, r = 0.6)

Arguments

Value

A data frame with three+n_vars columns, like:

x

Random uniform coordinate

y

Random uniform coordinate

class

Binary class label ("Above" or "Below")

Examples

coords <- matrix(c(0.2, 0.3,
                   0.4, 0.5,
                   0.6, 0.7), ncol = 2, byrow = TRUE)

df <- create_data(coord = coords, n_samples = 500, seed = 717)

df <- trim_shape(df)
plot(df$x, df$y, col=ifelse(df$class == "Above", "red", "yellow"))