Get the Data Type of an Array

Returns the data type of an array (e.g. f32, i64).

Usage

dtype(x, ...)

Arguments

x

(AnvlArray | GraphBox | AbstractArray)
An array. See Details for the values that have no data type.

...

Additional arguments passed to methods (unused).

Value

(DataType)

Details

An R value has no data type of its own: it only takes one when it meets a typed array, or when it materializes at the default. So dtype() is an error for a plain R value, and also for an RData, the AbstractArray of an R value passed to a jit-compiled function, and so for the GraphBox that carries one during tracing. Use peek_dtype() for the data type such a value would take.

This is implemented via the generic xlamisc::dtype().

See also

xlamisc::dtype(), peek_dtype(), RData

Examples

x <- nv_array(1:4, dtype = "f32")
dtype(x)
#> <f32>
# an R value passed to a jit-compiled function has no data type
try(jit(dtype)(1))
#> Error : An R value has no data type of its own until it is used.
#> ℹ `dtype()` is undefined here for the same reason `dtype(1.5)` is: the value
#>   only takes a data type when it meets a typed array, or when it materializes
#>   at the default.
#> ℹ Give it one explicitly with `nv_convert()`.
# an array passed to it does
jit(function(x) {
  print(dtype(x))
  x
})(nv_scalar(1))
#> <f32>
#> AnvlArray
#>  1
#> [ CPUf32{} ]