class Array(T)
- Array(T)
- Reference
- Object
Overview
An Array
is an ordered, integer-indexed collection of objects of type T.
Array indexing starts at 0. A negative index is assumed to be relative to the end of the array: -1 indicates the last element, -2 is the next to last element, and so on.
An Array
can be created using the usual new
method (several are provided), or with an array literal:
Array(Int32).new # => []
[1, 2, 3] # Array(Int32)
[1, "hello", 'x'] # Array(Int32 | String | Char)
See Array
literals in the language reference.
An Array
can have mixed types, meaning T will be a union of types, but these are determined
when the array is created, either by specifying T or by using an array literal. In the latter
case, T will be set to the union of the array literal elements' types.
When creating an empty array you must always specify T:
[] of Int32 # same as Array(Int32)
[] # syntax error
An Array
is implemented using an internal buffer of some capacity
and is reallocated when elements are pushed to it when more capacity
is needed. This is normally known as a dynamic array.
You can use a special array literal syntax with other types too, as long as they define an argless
new
method and a <<
method. Set
is one such type:
set = Set{1, 2, 3} # => Set{1, 2, 3}
set.class # => Set(Int32)
The above is the same as this:
set = Set(typeof(1, 2, 3)).new
set << 1
set << 2
set << 3
Included Modules
- Comparable(Array(T))
- Indexable::Mutable(T)
Defined in:
extensions/array.crInstance Method Summary
-
#to_tensor(device = CPU)
Converts a standard library array of any shape to a
Tensor
.
Instance Method Detail
Converts a standard library array of any shape to a Tensor
.
The array, if nested, must have a uniform shape in all dimensions,
and each innermost element must have the same data type.
Arguments
- device :
Num::Storage
- The storage backend on which to place theTensor
Examples
a = [1, 2, 3]
typeof(a.to_tensor) # => Tensor(Int32, CPU(Float32))