Skip to contents

Plot the robust regression or explanation from slise

Usage

# S3 method for slise
plot(x, type = NULL, title = NULL, ...)

Arguments

x

The slise object

type

The type of plot ("2D", "bar", "distribution", "mnist", "prediction", "wordcloud")

title

The title of the plot (may include a `%s`, which will be replaced by the prediction)

...

Arguments passed on to plot.slise_2d, plot.slise_bar, plot.slise_distribution, plot.slise_mnist, plot.slise_prediction, plot.slise_wordcloud

labels

The axis labels (default: c("X", "Y") or c("x", "f(x)"))

partial

Should the raw ggplot2 objects be returned instead of directly plotting (default: FALSE)

size

The size of the plot elements (default: 2)

signif

The number of significant digits to display (default: 3)

width

The width of the image (width * height == ncol(X))

height

The height of the image (width * height == ncol(X))

plots

The number of plots to split the explanation into (default: 1)

treshold

Treshold for ignored value (default: 1e-8)

local

Only display the words relevant for the explained item (default: TRUE)

Value

plot or ggplot2 objects

Examples

X <- matrix(rnorm(30), 30, 1)
Y <- runif(30, 0, 1)
plot(slise.fit(X, Y, epsilon = 0.1))
#> Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
#> of ggplot2 3.3.4.
#>  The deprecated feature was likely used in the slise package.
#>   Please report the issue to the authors.