Skip to contents

Median

Source: R/api.R, R/api-generics.R
nv_median.Rd

Computes the median over one or more axes. Equivalent to nv_quantile(x, 0.5, axes, drop, interpolation); for an even number of reduced elements with the default "linear" interpolation, the average of the two middle values is returned, matching base R's median().

You can also use median() directly on an AnvlArray or AnvlBox; extra arguments (e.g. interpolation) are forwarded via ....

Usage

nv_median(
  x,
  axes = NULL,
  drop = TRUE,
  interpolation = "linear",
  nan_rm = FALSE
)

# S3 method for class 'AnvlArray'
median(
  x,
  na.rm = FALSE,
  ...,
  axes = NULL,
  drop = TRUE,
  interpolation = "linear"
)

Arguments

x

(arrayish)
One input. Can be any data type. An R value materializes at its default data type.

axes

(integer() | NULL)
Axes to reduce over. Negative values count from the end, i.e. -1 refers to the last axis. If NULL (default), reduces over all axes.

drop

(logical(1))
Whether to drop the reduced axes: removed from the output shape if TRUE, set to 1 if FALSE.

interpolation

(character(1))
Forwarded to nv_quantile(). One of "linear" (default), "lower", "higher", "nearest", "midpoint".

nan_rm

(logical(1))
Forwarded to nv_quantile(). See its documentation for details.

na.rm

Forwarded to nv_median()'s nan_rm argument.

...

No additional arguments.

Value

(arrayish)
Same shape as x with axes removed (or set to 1 if drop = FALSE). The data type is that of x, or the default float for a non-float x.

The median() generic

stats::median() reduces every axis of a multi-axis array, and so does nv_median() by default, so the two agree. Pass axes to reduce a subset instead. A non-float x is computed at the default float, like base R returns a double.

See also

nv_quantile(), nv_sort(), prim_sort().

Examples

nv_median(nv_array(c(3, 1, 4, 1, 5, 9, 2, 6)))
#> AnvlArray
#>  3.5000
#> [ CPUf32{} ] 
median(nv_array(c(3, 1, 4, 1, 5, 9, 2, 6)))
#> AnvlArray
#>  3.5000
#> [ CPUf32{} ] 
m <- nv_matrix(c(3, 1, 5, 2, 4, 0), nrow = 2, byrow = TRUE)
nv_median(m) # over every element
#> AnvlArray
#>  2.5000
#> [ CPUf32{} ] 
nv_median(m, axes = 2L) # one median per row
#> AnvlArray
#>  3
#>  2
#> [ CPUf32{2} ] 
# forwards through the S3 generic via `...`
median(nv_array(c(1, 2, 3, 4)), interpolation = "lower")
#> AnvlArray
#>  2
#> [ CPUf32{} ] 
nv_median(nv_array(c(1, NaN, 3, 5)))
#> AnvlArray
#>  nan
#> [ CPUf32{} ] 
nv_median(nv_array(c(1, NaN, 3, 5)), nan_rm = TRUE)
#> AnvlArray
#>  3
#> [ CPUf32{} ]