Class MatchboxRecommenderPredictionSettings

Settings of the Matchbox recommender which affect prediction.

Inheritance

Object

MatchboxRecommenderPredictionSettings

Implements

ICustomSerializable

Inherited Members

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: Microsoft.ML.Probabilistic.Learners

Assembly: Microsoft.ML.Probabilistic.Learners.Recommender.dll

Syntax

[Serializable]
public class MatchboxRecommenderPredictionSettings : ICustomSerializable

Remarks

These settings can be modified after training.

Constructors

MatchboxRecommenderPredictionSettings()

Initializes a new instance of the MatchboxRecommenderPredictionSettings class.

Declaration

public MatchboxRecommenderPredictionSettings()

MatchboxRecommenderPredictionSettings(IReader)

Initializes a new instance of the MatchboxRecommenderPredictionSettings class from a reader of a binary stream.

Declaration

public MatchboxRecommenderPredictionSettings(IReader reader)

Parameters

Type	Name	Description
IReader	reader	The binary reader to read the prediction settings from.

Fields

LossFunctionDefault

The default loss function.

Declaration

public const LossFunction LossFunctionDefault = LossFunction.ZeroOne

Field Value

Type	Description
LossFunction

Methods

GetPredictionLossFunction(out Func<Int32, Int32, Double>)

Gets the loss function which determines how a prediction in the form of a distribution is converted into a point prediction.

Declaration

public LossFunction GetPredictionLossFunction(out Func<int, int, double> customLossFunction)

Parameters

Type	Name	Description
Func<Int32, Int32, Double>	customLossFunction	The custom loss function on integers. This is null unless the returned LossFunction is 'Custom'.

Returns

Type	Description
LossFunction	The LossFunction.

Remarks

A loss function returns the loss incurred when choosing an estimate instead of the true value, where the first argument is the true value and the second argument is the estimate of the true value.

SaveForwardCompatible(IWriter)

Saves the prediction settings of the Matchbox recommender using the specified writer to a binary stream.

Declaration

public void SaveForwardCompatible(IWriter writer)

Parameters

Type	Name	Description
IWriter	writer	The writer to save the prediction settings to.

SetPredictionLossFunction(LossFunction, Func<Int32, Int32, Double>)

Sets the loss function which determines how a prediction in the form of a distribution is converted into a point prediction.

Declaration

public void SetPredictionLossFunction(LossFunction lossFunction, Func<int, int, double> customLossFunction = null)

Parameters

Type	Name	Description
LossFunction	lossFunction	The loss function.
Func<Int32, Int32, Double>	customLossFunction	An optional custom loss function. This can only be set when lossFunction is set to 'Custom'. The custom loss function returns the loss incurred when choosing an estimate instead of the true value, where the first argument is the true value and the second argument is the estimate of the true value.

Implements

ICustomSerializable