ILNumerics Arrays
Whenever this documentation refers to 'arrays', it means the ILNumerics array from the core part. These computational arrays are used from any function of ILNumerics
and are the base of all algorithms. Those arrays may contain any number of dimensions and therefore reflect scalars, vectors, matrices and even n-dimensional tensors.
The following table lists some examples for common arrays and a way of creating them. Like almost all examples in this documentation, we expect
the code to be defined in the context of a class which is derived from
| array | creation examples |
|---|---|
|
matrix of size 4x3 element type: System.Double |
|
|
row vectorof length 100 element type: System.Int16 |
|
|
square matrix of size 10x10 element type: System.Single element values: 1.0f |
|
|
3-dimens.array of size 100x200x4 element type: double values: all |
A = array(pi,100,200,4) |
|
4-dimens.array of size 5x4x3x4 element type: double values counting from 1...240 |
|
The inner type T for
The most common way to create such arrays is by utilizing a matching initialization function from the
ILArray<T> is a mutable type. Once created, it can be altered in various ways easily. The spectrum ranges from simple value changes to size expansion and reshapes. Details are found in the Altering Arrays section.
ILArray<T> also plays an important role for the memory management of ILNumerics. By defining distinct array types for parameter lists in functions, ILNumerics is able to drastically optimize memory usage. The technique is described in detail in the General Rules section.
General Rules for ILNumericsAll array topics: