Plot a univariate or bivariate histogram

Port of seaborn.histplot. Returns a reaborn_plot.

Usage

histplot(
  data = NULL,
  x = NULL,
  y = NULL,
  hue = NULL,
  weights = NULL,
  stat = "count",
  bins = "auto",
  binwidth = NULL,
  binrange = NULL,
  discrete = NULL,
  cumulative = FALSE,
  common_bins = TRUE,
  common_norm = TRUE,
  multiple = "layer",
  element = "bars",
  fill = TRUE,
  shrink = 1,
  kde = FALSE,
  kde_kws = NULL,
  thresh = 0,
  cbar = FALSE,
  cbar_kws = NULL,
  palette = NULL,
  hue_order = NULL,
  hue_norm = NULL,
  color = NULL,
  cmap = NULL,
  legend = TRUE,
  .facet_vars = NULL,
  ...
)

Arguments

data

A data frame.

x, y

Column name/vector for the histogram variable (use y for a horizontal histogram).

hue

Grouping variable for color.

weights

Optional observation weights.

stat

One of "count", "frequency", "density", "probability", "proportion", "percent".

bins, binwidth, binrange, discrete

Binning controls (see rb_hist_bins).

cumulative

Accumulate counts.

common_bins, common_norm

Share bins/normalization across hue groups.

multiple

"layer", "stack", "fill", or "dodge".

element

"bars" or "step".

fill

Whether to fill the bars.

shrink

Shrink bar widths by this factor.

kde

Overlay a KDE curve.

kde_kws

Arguments for the KDE (e.g. bw_adjust).

thresh

Bivariate-only. Cells with a value at or below thresh are left transparent (default 0, so empty cells are blank); NULL fills every cell.

cbar, cbar_kws

Bivariate-only. Draw a color bar for the counts; cbar_kws accepts width (the bar width in points).

palette, hue_order, hue_norm, color

Color controls.

cmap

Bivariate-only colormap (name, reaborn_cmap, or color vector); defaults to a light sequential ramp built from color.

legend

Show the legend.

.facet_vars

Internal; facet columns forwarded by the figure-level dispatchers (catplot/displot/relplot). Not intended for direct use.

...

Passed to the bar geom.

Value

A reaborn_plot.

Details

When both x and y are supplied, histplot() draws a single 2-D count heatmap (like seaborn.histplot(x, y)). In that bivariate case hue is ignored, with a warning, and the hue-based color controls (palette, hue_order, hue_norm) do not apply; the fill is driven by cmap, which defaults to a light sequential ramp built from color.

Examples

penguins <- load_dataset("penguins")
histplot(data = penguins, x = "flipper_length_mm", hue = "species")


# Stack the hue groups
histplot(data = penguins, x = "flipper_length_mm", hue = "species", multiple = "stack")