ISmRandomForest.FilledDependent

Syntax

FilledDependent: Array;

Description

The FilledDependent property returns predicted classification.

Comments

To get summary results of classification, use the ISmRandomForest.ClassificationSummary property.

Example

Add a link to the Stat system assembly.

Sub UserProc;
Var
    rf: SmRandomForest;
    Y, CatX: Array[16] Of Integer;
    X: Array[16] Of Double;
    res, i, j: Integer;
    s: string;
    Explanatories: ISlSeries;
    Explan: ISlSerie;
    ExplInt: ISlSeriesInt;
    ExplanInt: ISlSerieInt;
    CatList: Array Of Integer;
Begin
    rf := New SmRandomForest.Create;
    // Set explained series
    Y[00] := 0; Y[08] := 1;
    Y[01] := 1; Y[09] := 2;
    Y[02] := 1; Y[10] := 0;
    Y[03] := 0; Y[11] := 2;
    Y[04] := 1; Y[12] := 1;
    Y[05] := 2; Y[13] := 2;
    Y[06] := 1; Y[14] := 2;
    Y[07] := 2; Y[15] := 0;
    rf.Dependent.Value := Y;
    // Set explanatory series (categorical)
    CatX[00] := 1; CatX[08] := 1;
    CatX[01] := 3; CatX[09] := 2;
    CatX[02] := 1; CatX[10] := 3;
    CatX[03] := 3; CatX[11] := 2;
    CatX[04] := 1; CatX[12] := 3;
    CatX[05] := 2; CatX[13] := 1;
    CatX[06] := 1; CatX[14] := 1;
    CatX[07] := 2; CatX[15] := 3;
    ExplInt := rf.ExplanatoriesCategorical;
    ExplInt.Clear;
    ExplanInt := ExplInt.Add;
    ExplanInt.Id := "Categorical_X";
    ExplanInt.Name := "CatX";
    ExplanInt.Value := CatX;
    // Set explanatory series (quantitative)
    X[00] := 34.13; X[08] := 29.27;
    X[01] := 21.52; X[09] := 23.39;
    X[02] := 25.43; X[10] := 28.28;
    X[03] := 43.42; X[11] := 43.55;
    X[04] := 40.19; X[12] := 44.80;
    X[05] := 24.97; X[13] := 23.23;
    X[06] := 20.57; X[14] := 37.14;
    X[07] := 30.81; X[15] := 27.44;
    Explanatories := rf.ExplanatoriesContinuous;
    Explanatories.Clear;
    Explan := Explanatories.Add;
    Explan.Id := "Continuous_X";
    Explan.Name := "X";
    Explan.Value := X;
    // Forest size
    rf.ForestSize := 20;
    rf.LearningSamplePortion := 0.6;
    // Trees size
    res := rf.Execute;
    If (res = 0) Then
        Debug.WriteLine(" === Probability  ===");
        Debug.Indent;
        For i := 0 To rf.Probability.GetUpperBound(1) Do
            For j := 0 To rf.Probability.GetUpperBound(2) Do
                s := s + rf.Probability[i, j].ToString + " ";
            End For;
            Debug.WriteLine(s);
            s := "";
        End For;
        Debug.Unindent;
        Debug.WriteLine(" === Predicted classification ===");
        Debug.Indent;
        For i := 0 To rf.FilledDependent.Length - 1 Do
            Debug.WriteLine((i + 1).ToString + ". " + rf.FilledDependent[i].ToString);
        End For;
        Debug.Unindent;
        Debug.WriteLine("=== Original values of categorical data series ===");
        Debug.Indent;
        For i := 0 To ExplanInt.Value.Length - 1 Do
            Debug.WriteLine((i + 1).ToString + ". " + ExplanInt.OriginalValue[i].ToString);
        End For;
        Debug.Unindent;
        Debug.WriteLine(" === Summary results of classification  ===");
        Debug.Indent;
        s := "";
        For i := 0 To rf.ClassificationSummary.GetUpperBound(1) Do
            For j := 0 To rf.ClassificationSummary.GetUpperBound(2) Do
                s := s + rf.ClassificationSummary[i, j].ToString + " ";
            End For;
            Debug.WriteLine(s);
            s := "";
        End For;
        Debug.Unindent;
        // Display list of categories
        CatList := rf.CategoriesList;
        If CatList.Length > 0 Then
            Debug.WriteLine("List of categories:"); Debug.Indent;
            For i := 0 To CatList.Length - 1 Do
                Debug.WriteLine(CatList[i]);
            End For;
            Debug.Unindent;
        End If;
        Else
            Debug.WriteLine(rf.Errors);
    End If;
End Sub UserProc;

After executing the example the console window displays:

Probability.
Predicted classification.
Original values of categorical data series.
Summary results of classification.
List of categories.