Package org.flag4j.operations.dense.real

Class RealDenseElemMult

java.lang.Object

org.flag4j.operations.dense.real.RealDenseElemMult

public class RealDenseElemMult
extends Object

This class contains low level implementations of element-wise multiplications algorithms for real dense tensors.

Field Summary

Fields

Modifier and Type	Field	Description
private static final int	CONCURRENT_THRESHOLD	Minimum number of entries in each tensor to apply concurrent algorithm.

Constructor Summary

Constructors

Modifier	Constructor	Description
private	RealDenseElemMult()

Method Summary

Modifier and Type	Method	Description
static double[]	dispatch(double[] src1, Shape shape1, double[] src2, Shape shape2)	Dynamically chooses and applies element-wise multiplication algorithm to use based on the number of entries in the tensors.
static double[]	elemMult(double[] src1, Shape shape1, double[] src2, Shape shape2)	Computes the element-wise multiplication of two tensors.
static double[]	elemMultConcurrent(double[] src1, Shape shape1, double[] src2, Shape shape2)	Computes the element-wise multiplication of two tensors using a concurrent algorithm.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

CONCURRENT_THRESHOLD

private static final int CONCURRENT_THRESHOLD

Minimum number of entries in each tensor to apply concurrent algorithm.

See Also:

• Constant Field Values

Constructor Details

RealDenseElemMult

private RealDenseElemMult()

Method Details

elemMult

public static double[] elemMult(double[] src1,
 Shape shape1,
 double[] src2,
 Shape shape2)

Computes the element-wise multiplication of two tensors. Also called the Hadamard product.

Parameters:

src1 - First tensor in element-wise multiplication.

shape1 - Shape of the first tensor.

src2 - Second tensor in element-wise multiplication.

shape2 - Shape of the second tensor.

Returns:

The element-wise multiplication of the two tensors.

Throws:

IllegalArgumentException - If the tensors do not have the same shape.

elemMultConcurrent

public static double[] elemMultConcurrent(double[] src1,
 Shape shape1,
 double[] src2,
 Shape shape2)

Computes the element-wise multiplication of two tensors using a concurrent algorithm. Also called the Hadamard product.

Parameters:

src1 - First tensor in element-wise multiplication.

shape1 - Shape of the first tensor.

src2 - Second tensor in element-wise multiplication.

shape2 - Shape of the second tensor.

Returns:

The element-wise multiplication of the two tensors.

Throws:

IllegalArgumentException - If the tensors do not have the same shape.

dispatch

public static double[] dispatch(double[] src1,
 Shape shape1,
 double[] src2,
 Shape shape2)

Dynamically chooses and applies element-wise multiplication algorithm to use based on the number of entries in the tensors.

Parameters:

src1 - Entries of first tensor.

shape1 - Shape of first tensor.

src2 - Entries of second tensor.

shape2 - Shape of second tensor.

Returns:

The element-wise multiplication of the two tensors.