MatrixFactorizationTrainer+Options.xml

<Type Name="MatrixFactorizationTrainer+Options" FullName="Microsoft.ML.Trainers.MatrixFactorizationTrainer+Options">
  <TypeSignature Language="C#" Value="public sealed class MatrixFactorizationTrainer.Options" />
  <TypeSignature Language="ILAsm" Value=".class nested public auto ansi sealed beforefieldinit MatrixFactorizationTrainer/Options extends System.Object" />
  <TypeSignature Language="DocId" Value="T:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options" />
  <TypeSignature Language="VB.NET" Value="Public NotInheritable Class MatrixFactorizationTrainer.Options" />
  <TypeSignature Language="F#" Value="type MatrixFactorizationTrainer.Options = class" />
  <AssemblyInfo>
    <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
    <AssemblyVersion>1.0.0.0</AssemblyVersion>
  </AssemblyInfo>
  <Base>
    <BaseTypeName>System.Object</BaseTypeName>
  </Base>
  <Interfaces />
  <Docs>
    <summary>
            Options for the <see cref="T:Microsoft.ML.Trainers.MatrixFactorizationTrainer" /> as used in [MatrixFactorization(Options)](xref:Microsoft.ML.RecommendationCatalog.RecommendationTrainers.MatrixFactorization(Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options)).
            </summary>
    <remarks>To be added.</remarks>
  </Docs>
  <Members>
    <Member MemberName=".ctor">
      <MemberSignature Language="C#" Value="public Options ();" />
      <MemberSignature Language="ILAsm" Value=".method public hidebysig specialname rtspecialname instance void .ctor() cil managed" />
      <MemberSignature Language="DocId" Value="M:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.#ctor" />
      <MemberSignature Language="VB.NET" Value="Public Sub New ()" />
      <MemberType>Constructor</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <Parameters />
      <Docs>
        <summary>To be added.</summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="Alpha">
      <MemberSignature Language="C#" Value="public double Alpha;" />
      <MemberSignature Language="ILAsm" Value=".field public float64 Alpha" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Alpha" />
      <MemberSignature Language="VB.NET" Value="Public Alpha As Double " />
      <MemberSignature Language="F#" Value="val mutable Alpha : double" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Alpha" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Double</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Importance of unobserved entries' loss in one-class matrix factorization. Applicable if <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LossFunction" /> set to <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossOneClass" /></summary>
        <remarks>
            Importance of unobserved (i.e., negative) entries' loss in one-class matrix factorization.
            In general, only a few of matrix entries (e.g., less than 1%) in the training are observed (i.e., positive).
            To balance the contributions from unobserved and observed in the overall loss function, this parameter is
            usually a small value so that the solver is able to find a factorization equally good to unobserved and observed
            entries. If only 10000 observed entries present in a 200000-by-300000 training matrix, one can try Alpha = 10000 / (200000*300000 - 10000).
            When most entries in the training matrix are observed, one can use Alpha &gt;&gt; 1; for example, if only 10000 in previous
            matrix is not observed, one can try Alpha = (200000 * 300000 - 10000) / 10000. Consequently,
            Alpha = (# of observed entries) / (# of unobserved entries) can make observed and unobserved entries equally important
            in the minimized loss function. However, the best setting in machine learning is always data-dependent so user still needs to
            try multiple values.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="ApproximationRank">
      <MemberSignature Language="C#" Value="public int ApproximationRank;" />
      <MemberSignature Language="ILAsm" Value=".field public int32 ApproximationRank" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.ApproximationRank" />
      <MemberSignature Language="VB.NET" Value="Public ApproximationRank As Integer " />
      <MemberSignature Language="F#" Value="val mutable ApproximationRank : int" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.ApproximationRank" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Int32</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Rank of approximation matrices.
            </summary>
        <remarks>
            If input data has size of m-by-n we would build two approximation matrices m-by-k and k-by-n where k is approximation rank.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="C">
      <MemberSignature Language="C#" Value="public double C;" />
      <MemberSignature Language="ILAsm" Value=".field public float64 C" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.C" />
      <MemberSignature Language="VB.NET" Value="Public C As Double " />
      <MemberSignature Language="F#" Value="val mutable C : double" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.C" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Double</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Desired negative entries value in one-class matrix factorization. Applicable if <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LossFunction" /> set to <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossOneClass" /></summary>
        <remarks>
            In one-class matrix factorization, all matrix values observed are one (which can be viewed as positive cases in binary classification)
            while unobserved values (which can be viewed as negative cases in binary classification) need to be specified manually using this option.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="LabelColumnName">
      <MemberSignature Language="C#" Value="public string LabelColumnName;" />
      <MemberSignature Language="ILAsm" Value=".field public string LabelColumnName" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LabelColumnName" />
      <MemberSignature Language="VB.NET" Value="Public LabelColumnName As String " />
      <MemberSignature Language="F#" Value="val mutable LabelColumnName : string" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LabelColumnName" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.String</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            The name variable (i.e., column in a <see cref="T:Microsoft.ML.IDataView" /> type system) used as matrix's element value.
            The column data must be <see cref="T:Microsoft.ML.Data.KeyDataViewType" />.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="Lambda">
      <MemberSignature Language="C#" Value="public double Lambda;" />
      <MemberSignature Language="ILAsm" Value=".field public float64 Lambda" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Lambda" />
      <MemberSignature Language="VB.NET" Value="Public Lambda As Double " />
      <MemberSignature Language="F#" Value="val mutable Lambda : double" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Lambda" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Double</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Regularization parameter.
            </summary>
        <remarks>
            It's the weight of factor matrices Frobenius norms in the objective function minimized by matrix factorization's algorithm. A small value could cause over-fitting.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="LearningRate">
      <MemberSignature Language="C#" Value="public double LearningRate;" />
      <MemberSignature Language="ILAsm" Value=".field public float64 LearningRate" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LearningRate" />
      <MemberSignature Language="VB.NET" Value="Public LearningRate As Double " />
      <MemberSignature Language="F#" Value="val mutable LearningRate : double" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LearningRate" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Double</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
             Initial learning rate. It specifies the speed of the training algorithm.
            </summary>
        <remarks>
             Small value may increase the number of iterations needed to achieve a reasonable result.
             Large value may lead to numerical difficulty such as a infinity value.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="LossFunction">
      <MemberSignature Language="C#" Value="public Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType LossFunction;" />
      <MemberSignature Language="ILAsm" Value=".field public valuetype Microsoft.ML.Trainers.MatrixFactorizationTrainer/LossFunctionType LossFunction" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LossFunction" />
      <MemberSignature Language="VB.NET" Value="Public LossFunction As MatrixFactorizationTrainer.LossFunctionType " />
      <MemberSignature Language="F#" Value="val mutable LossFunction : Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.LossFunction" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>Microsoft.ML.Trainers.MatrixFactorizationTrainer+LossFunctionType</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Loss function minimized for finding factor matrices.
            </summary>
        <remarks>
            Two values are allowed, <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossRegression" /> or <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossOneClass" />.
            The <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossRegression" /> means traditional collaborative filtering problem with squared loss.
            The <see cref="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.LossFunctionType.SquareLossOneClass" /> triggers one-class matrix factorization for implicit-feedback recommendation problem.
            </remarks>
      </Docs>
    </Member>
    <Member MemberName="MatrixColumnIndexColumnName">
      <MemberSignature Language="C#" Value="public string MatrixColumnIndexColumnName;" />
      <MemberSignature Language="ILAsm" Value=".field public string MatrixColumnIndexColumnName" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.MatrixColumnIndexColumnName" />
      <MemberSignature Language="VB.NET" Value="Public MatrixColumnIndexColumnName As String " />
      <MemberSignature Language="F#" Value="val mutable MatrixColumnIndexColumnName : string" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.MatrixColumnIndexColumnName" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.String</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            The name of variable (i.e., Column in a <see cref="T:Microsoft.ML.IDataView" /> type system) used as matrix's column index.
            The column data must be <see cref="T:System.Single" />.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="MatrixRowIndexColumnName">
      <MemberSignature Language="C#" Value="public string MatrixRowIndexColumnName;" />
      <MemberSignature Language="ILAsm" Value=".field public string MatrixRowIndexColumnName" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.MatrixRowIndexColumnName" />
      <MemberSignature Language="VB.NET" Value="Public MatrixRowIndexColumnName As String " />
      <MemberSignature Language="F#" Value="val mutable MatrixRowIndexColumnName : string" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.MatrixRowIndexColumnName" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.String</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            The name of variable (i.e., column in a <see cref="T:Microsoft.ML.IDataView" /> type system) used as matrix's row index.
            The column data must be <see cref="T:Microsoft.ML.Data.KeyDataViewType" />.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="NonNegative">
      <MemberSignature Language="C#" Value="public bool NonNegative;" />
      <MemberSignature Language="ILAsm" Value=".field public bool NonNegative" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NonNegative" />
      <MemberSignature Language="VB.NET" Value="Public NonNegative As Boolean " />
      <MemberSignature Language="F#" Value="val mutable NonNegative : bool" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NonNegative" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Boolean</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Force the factor matrices to be non-negative.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="NumberOfIterations">
      <MemberSignature Language="C#" Value="public int NumberOfIterations;" />
      <MemberSignature Language="ILAsm" Value=".field public int32 NumberOfIterations" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NumberOfIterations" />
      <MemberSignature Language="VB.NET" Value="Public NumberOfIterations As Integer " />
      <MemberSignature Language="F#" Value="val mutable NumberOfIterations : int" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NumberOfIterations" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Int32</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Number of training iterations.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="NumberOfThreads">
      <MemberSignature Language="C#" Value="public int? NumberOfThreads;" />
      <MemberSignature Language="ILAsm" Value=".field public valuetype System.Nullable`1&lt;int32&gt; NumberOfThreads" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NumberOfThreads" />
      <MemberSignature Language="VB.NET" Value="Public NumberOfThreads As Nullable(Of Integer) " />
      <MemberSignature Language="F#" Value="val mutable NumberOfThreads : Nullable&lt;int&gt;" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.NumberOfThreads" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Nullable&lt;System.Int32&gt;</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Number of threads will be used during training. If unspecified all available threads will be use.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
    <Member MemberName="Quiet">
      <MemberSignature Language="C#" Value="public bool Quiet;" />
      <MemberSignature Language="ILAsm" Value=".field public bool Quiet" />
      <MemberSignature Language="DocId" Value="F:Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Quiet" />
      <MemberSignature Language="VB.NET" Value="Public Quiet As Boolean " />
      <MemberSignature Language="F#" Value="val mutable Quiet : bool" Usage="Microsoft.ML.Trainers.MatrixFactorizationTrainer.Options.Quiet" />
      <MemberType>Field</MemberType>
      <AssemblyInfo>
        <AssemblyName>Microsoft.ML.Recommender</AssemblyName>
        <AssemblyVersion>1.0.0.0</AssemblyVersion>
      </AssemblyInfo>
      <ReturnValue>
        <ReturnType>System.Boolean</ReturnType>
      </ReturnValue>
      <Docs>
        <summary>
            Suppress writing additional information to output.
            </summary>
        <remarks>To be added.</remarks>
      </Docs>
    </Member>
  </Members>
</Type>