Skip to contents

Create an Array of a Given Data Type

Source: R/array.R
nv_dtype_constructors.Rd

Shorthands for nv_array() with the data type fixed by the function name: nv_float32(data) is nv_array(data, dtype = "f32").

FunctionData type
nv_bool()bool
nv_int8()i8
nv_int16()i16
nv_int32()i32
nv_int64()i64
nv_uint8()ui8
nv_uint16()ui16
nv_uint32()ui32
nv_uint64()ui64
nv_float32()f32
nv_float64()f64

Usage

nv_bool(data, shape = NULL, device = NULL, byrow = FALSE)

nv_int8(data, shape = NULL, device = NULL, byrow = FALSE)

nv_int16(data, shape = NULL, device = NULL, byrow = FALSE)

nv_int32(data, shape = NULL, device = NULL, byrow = FALSE)

nv_int64(data, shape = NULL, device = NULL, byrow = FALSE)

nv_uint8(data, shape = NULL, device = NULL, byrow = FALSE)

nv_uint16(data, shape = NULL, device = NULL, byrow = FALSE)

nv_uint32(data, shape = NULL, device = NULL, byrow = FALSE)

nv_uint64(data, shape = NULL, device = NULL, byrow = FALSE)

nv_float32(data, shape = NULL, device = NULL, byrow = FALSE)

nv_float64(data, shape = NULL, device = NULL, byrow = FALSE)

Arguments

data

(any)
integer(), double(), or logical() scalar, vector, or array. Alternatively a raw() vector holding the native little-endian byte payload of prod(shape) elements of dtype; both dtype and shape are then required (only supported on the "pjrt" backend). Raw payloads are read in column-major element order, or row-major with byrow = TRUE. An existing AnvlArray is returned unchanged if the shape, dtype and device given agree with it, and is an error otherwise.

shape

(NULL | integer())
The output shape of the array. The default (NULL) is to infer it from the data if possible. Note that nv_array interprets length 1 vectors as having shape (1). Empty data has no shape to infer – 0, c(2, 0) and c(0, 3) all hold no elements – so shape is required there. To create a "scalar" with no axes (shape ()), use nv_scalar or explicitly specify shape = integer().

device

(NULL | character(1) | device)
The device the data lives on, given either as:

  • a device string naming the platform (e.g. "cpu", "cuda", "cuda:<n>"), which is resolved against the backend in use, or

  • a device object as returned by nv_device(): a PJRTDevice for the "pjrt" backend or a quickr_device for the "quickr" backend. It must belong to the active backend (active_backend()); a device of another backend is an error.

The default (NULL) uses default_device().

byrow

(logical(1))
When constructing from an R object and the result has at least two axes, fill the array in row-major order rather than the default column-major order. Only allowed when data is an R object — passing an existing AnvlArray together with byrow = TRUE is an error.

Value

(AnvlArray)

Out of Range values

Because base R has fewer data types than anvl, creating AnvlArrays from R often involves type conversions. When such conversions are performed, anvl performs a scan of the inputs to ensure that the requested data type can actually hold the input data. For example, trying to create an unsigned integer from a negative R integer() fails. The same holds where an R value takes its data type from the array it meets rather than from an argument: nv_scalar(1L, "ui8") + (-2L) is refused, where converting an array with nv_convert() wraps around.

See also

nv_array(), nv_convert() to change the data type of an existing array.

Examples

nv_float32(1:4)
#> AnvlArray
#>  1
#>  2
#>  3
#>  4
#> [ CPUf32{4} ] 
nv_uint8(c(0, 255))
#> AnvlArray
#>    0
#>  255
#> [ CPUui8{2} ] 
nv_int64(1:6, shape = c(2, 3))
#> AnvlArray
#>  1 3 5
#>  2 4 6
#> [ CPUi64{2,3} ] 
nv_bool(c(TRUE, FALSE))
#> AnvlArray
#>  1
#>  0
#> [ CPUbool{2} ]