Class Sparsity

Defines sparsity settings for vectors. The sparsity handling has been designed to deal with large dimensional distributions such as Discrete and Dirichlet distributions.

Inheritance

Object

Sparsity

Inherited Members

Object.Equals(Object, Object)

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Namespace: Microsoft.ML.Probabilistic.Math

Assembly: Microsoft.ML.Probabilistic.dll

Syntax

[Serializable]
[DataContract]
public class Sparsity

Properties

CountTolerance

The maximum allowed count of vector elements not set to the common value.

Declaration

[DataMember]
public int CountTolerance { get; }

Property Value

Type	Description
Int32

Dense

Static read-only property giving Sparsity settings for dense vectors.

Declaration

public static Sparsity Dense { get; }

Property Value

Type	Description
Sparsity

Remarks

DenseVector uses this specification.

IsApproximate

True if the sparsity is approximate

Declaration

public bool IsApproximate { get; }

Property Value

Type	Description
Boolean

IsDense

True if is the vector is dense

Declaration

public bool IsDense { get; }

Property Value

Type	Description
Boolean

IsExact

True if the sparsity is exact

Declaration

public bool IsExact { get; }

Property Value

Type	Description
Boolean

IsPiecewise

True if is the vector is piecewise

Declaration

public bool IsPiecewise { get; }

Property Value

Type	Description
Boolean

IsSparse

True if the vector is sparse (exact or approximate)

Declaration

public bool IsSparse { get; }

Property Value

Type	Description
Boolean

Piecewise

Static read-only property giving Sparsity settings for piecewise, exact vectors.

Declaration

public static Sparsity Piecewise { get; }

Property Value

Type	Description
Sparsity

Remarks

A vector class which uses this sparsity is expected to have ranges of elements that have the same values. PiecewiseVector uses this specification.

Sparse

Static read-only property giving Sparsity settings for sparse, exact vectors.

Declaration

public static Sparsity Sparse { get; }

Property Value

Type	Description
Sparsity

Remarks

A vector class which uses this sparsity will maintain a common background value. It is expected that only a small percentage of values will differ from this common value, and this enables the vector class to be efficient computationally and memory-wise. SparseVector uses this specification.

Storage

The storage method used by this vector

Declaration

public StorageType Storage { get; }

Property Value

Type	Description
StorageType

Tolerance

The tolerance at which vector element values are considered equal to the common value.

Declaration

[DataMember]
public double Tolerance { get; }

Property Value

Type	Description
Double

Methods

ApproximateWithTolerance(Double)

Static read-only property giving Sparsity settings for approximate vector with the given tolerance.

Declaration

public static Sparsity ApproximateWithTolerance(double tolerance)

Parameters

Type	Name	Description
Double	tolerance	At what tolerance are vector values considered equal to the common value.

Returns

Type	Description
Sparsity

Remarks

ApproximateSparseVector supports this specification.

Equals(Object)

Determines the equality of this instance with another

Declaration

public override bool Equals(object obj)

Parameters

Type	Name	Description
Object	obj

Returns

Type	Description
Boolean

Overrides

Object.Equals(Object)

FromSpec(StorageType, Double, Int32)

Static method to create a general sparsity specification

Declaration

[Construction(new string[]{"Storage", "Tolerance", "CountTolerance"})]
public static Sparsity FromSpec(StorageType storage, double tolerance, int countTolerance)

Parameters

Type	Name	Description
StorageType	storage	The storage method the vector should use
Double	tolerance	The tolerance at which vector element values are considered equal to the common value.
Int32	countTolerance	The maximum allowed count of vector elements not set to the common value.

Returns

Type	Description
Sparsity

GetHashCode()

Declaration

public override int GetHashCode()

Returns

Type	Description
Int32

Overrides

Object.GetHashCode()

ToString()

Instance description

Declaration

public override string ToString()

Returns

Type	Description
String

Overrides

Object.ToString()

Operators

Equality(Sparsity, Sparsity)

Equality operator.

Declaration

public static bool operator ==(Sparsity a, Sparsity b)

Parameters

Type	Name	Description
Sparsity	a	First sparsity.
Sparsity	b	Second sparsity.

Returns

Type	Description
Boolean	True if the sparsity specifications are equal.

Inequality(Sparsity, Sparsity)

Inequality operator.

Declaration

public static bool operator !=(Sparsity a, Sparsity b)

Parameters

Type	Name	Description
Sparsity	a	First sparsity specification.
Sparsity	b	Second sparsity specification.

Returns

Type	Description
Boolean	True if sparsity specifications are not equal.