org.tensorics.core.tensor

Class BroadcastedTensorView<V>

java.lang.Object

org.tensorics.core.tensor.AbstractTensor<V>

org.tensorics.core.tensor.BroadcastedTensorView<V>

Type Parameters:

V - the type of the values of the tensor

All Implemented Interfaces:

Tensor<V>, Tensoric<V>

public final class BroadcastedTensorView<V>
extends AbstractTensor<V>

Lets a tensors appear as a tensor with a bigger shape. The final ('broadcasted') shape of the tensor has more dimensions as the original tensor and is the outer product of the original shape and an additional shape passed into the view. The original set and the passed in shape have to be disjunct in dimensions, i.e. there must be no dimension which appears in both of them.

The values are broadcasted such that the values will be the same for all coordinates of the new dimensions.

The resulting view will still be backed by the original tensor. Since the broadcasted shape will never be updated from the original tensor, the results might be unpredictable, if the original tensor would be mutable.

Constructor Summary

Constructors

Constructor and Description
BroadcastedTensorView(Tensor<V> originalTensor, Shape extendingShape) Constructs a view of the given original tensor, broadcasted to the additional shape.

Method Summary

Modifier and Type	Method and Description
boolean	contains(Position position) Returns true if the tensoric object contains the given position, false otherwise.
Position	context() Retrieves the context of the tensor, which is nothing else than a position.
V	get(java.lang.Object... coordinates)
V	get(Position position)
Shape	shape() Retrieves the shape of the tensor.
java.lang.String	toString()

Methods inherited from class org.tensorics.core.tensor.AbstractTensor

equals, hashCode

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.tensorics.core.tensor.Tensoric

contains

Constructor Detail

BroadcastedTensorView

public BroadcastedTensorView(Tensor<V> originalTensor,
                             Shape extendingShape)

Constructs a view of the given original tensor, broadcasted to the additional shape.

Parameters:

originalTensor - the original tensor

extendingShape - the shape by which the original tensor shape has to be enlarged

Method Detail

get

public V get(Position position)

Parameters:

position - the position in the N-dimensional space where to find the value.

Returns:

the value at the given position

get

public V get(java.lang.Object... coordinates)

Parameters:

coordinates - form N-dimensional space where to find the value.

Returns:

a value at the given coordinates.

shape

public Shape shape()

Description copied from interface: Tensor

Retrieves the shape of the tensor. As shape we understand simply the structure of a tensor: Its dimensions and the available positions.

Implementations have to take care that the returned value here is never null.

Returns:

the shape of the tensor.

context

public Position context()

Description copied from interface: Tensor

Retrieves the context of the tensor, which is nothing else than a position. As context of the tensor we understand coordinates within a higher dimensional space than than the tensor has itself. This coordinates can e.g. transport information when e.g. a higher-dimensional tensor is split into smaller ones, so that it can potentially be reconstructed afterwards.

Implementations have to guarantee that the returned value here is never null.

Returns:

the context of the tensor.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

contains

public boolean contains(Position position)

Description copied from interface: Tensoric

Returns true if the tensoric object contains the given position, false otherwise.

Parameters:

position - the position to be checked

Returns:

true if the position is contained in the tensoric object, false otherwise.