Transpose

Permutes the axes of an array, like base::aperm(). nv_transpose() is another spelling of the same function. You can also use aperm(), or t() for matrices.

Usage

nv_aperm(x, perm = NULL)

nv_transpose(x, perm = NULL)

# S3 method for class 'AnvlArray'
t(x)

# S3 method for class 'AnvlArray'
aperm(a, perm = NULL, ...)

Arguments

x

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

perm

(integer() | NULL)
New ordering of axes. If NULL (default), reverses the axes. Negative values count from the end, i.e. -1 refers to the last axis.

a

(arrayish)
The array whose axes to permute.

...

No additional arguments.

Value

(arrayish)
Has x's data type and shape shape(x)[perm], or rev(shape(x)) when perm is NULL.

The t() generic

t() requires a matrix, whereas base::t() also transposes a vector (into a one-row matrix). Like base::t(), it rejects a higher-rank array; use aperm() to reverse its axes.

The aperm() generic

aperm() permutes the axes exactly as nv_aperm() does; like base::aperm(), perm = NULL reverses them.

See also

prim_transpose() for the underlying primitive.

Examples

# the 2x3 becomes a 3x2, keeping its data type
x <- nv_matrix(1:6, nrow = 2)
nv_aperm(x)
#> AnvlArray
#>  1 2
#>  3 4
#>  5 6
#> [ CPUi32{3,2} ] 
t(x)
#> AnvlArray
#>  1 2
#>  3 4
#>  5 6
#> [ CPUi32{3,2} ] 
nv_aperm(nv_array(1:24, shape = c(2, 3, 4)), c(3, 1, 2))
#> AnvlArray
#> (1,.,.) =
#>  1 3 5
#>  2 4 6
#> 
#> (2,.,.) =
#>   7  9 11
#>   8 10 12
#> 
#> (3,.,.) =
#>  13 15 17
#>  14 16 18
#> 
#> (4,.,.) =
#>  19 21 23
#>  20 22 24
#> [ CPUi32{4,2,3} ]