Search Results for

    Show / Hide Table of Contents

    Class MatchboxRecommenderTrainingSettings

    Settings of the Matchbox recommender which affect training. Cannot be set after training.

    Inheritance
    Object
    MatchboxRecommenderTrainingSettings
    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 MatchboxRecommenderTrainingSettings : ICustomSerializable

    Fields

    BatchCountDefault

    The default number of data batches.

    Declaration
    public const int BatchCountDefault = 1
    Field Value
    Type Description
    Int32

    IterationCountDefault

    The default number of inference iterations.

    Declaration
    public const int IterationCountDefault = 20
    Field Value
    Type Description
    Int32

    TraitCountDefault

    The default number of traits.

    Declaration
    public const int TraitCountDefault = 4
    Field Value
    Type Description
    Int32

    UseItemFeaturesDefault

    The default value indicating whether item features will be used.

    Declaration
    public const bool UseItemFeaturesDefault = false
    Field Value
    Type Description
    Boolean

    UseSharedUserThresholdsDefault

    The default value indicating whether shared user thresholds will be used.

    Declaration
    public const bool UseSharedUserThresholdsDefault = false
    Field Value
    Type Description
    Boolean

    UseUserFeaturesDefault

    The default value indicating whether user features will be used.

    Declaration
    public const bool UseUserFeaturesDefault = false
    Field Value
    Type Description
    Boolean

    Properties

    Advanced

    Gets the advanced settings of the Matchbox recommender.

    Declaration
    public MatchboxRecommenderAdvancedTrainingSettings Advanced { get; }
    Property Value
    Type Description
    MatchboxRecommenderAdvancedTrainingSettings

    BatchCount

    Gets or sets the number of batches the training data is split into.

    Declaration
    public int BatchCount { get; set; }
    Property Value
    Type Description
    Int32

    IterationCount

    Gets or sets the number of inference iterations to run.

    Declaration
    public int IterationCount { get; set; }
    Property Value
    Type Description
    Int32

    TraitCount

    Gets or sets the number of implicit user or item features (traits) to learn.

    Declaration
    public int TraitCount { get; set; }
    Property Value
    Type Description
    Int32

    UseItemFeatures

    Gets or sets a value indicating whether to use explicit item features.

    Declaration
    public bool UseItemFeatures { get; set; }
    Property Value
    Type Description
    Boolean

    UseSharedUserThresholds

    Gets or sets a value indicating whether to use shared user thresholds.

    Declaration
    public bool UseSharedUserThresholds { get; set; }
    Property Value
    Type Description
    Boolean

    UseUserFeatures

    Gets or sets a value indicating whether to use explicit user features.

    Declaration
    public bool UseUserFeatures { get; set; }
    Property Value
    Type Description
    Boolean

    Methods

    SaveForwardCompatible(IWriter)

    Saves the training 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 training settings to.

    Implements

    ICustomSerializable
    In This Article
    Back to top Copyright © .NET Foundation. All rights reserved.