RandomForest

public class RandomForest

Undocumented

  • Undocumented

    Declaration

    Swift

    public var forest: [RandomTree]

  • Undocumented

    Declaration

    Swift

    public var fullData: [[String]]

  • Undocumented

    Declaration

    Swift

    public var target: Int

  • Undocumented

    Declaration

    Swift

    public var nTrees: Int

  • Undocumented

    Declaration

    Swift

    public var nFeatures: Int

  • Undocumented

    Declaration

    Swift

    public var depth: Int

  • Undocumented

    Declaration

    Swift

    public var perform: String

  • Undocumented

    Declaration

    Swift

    public var using: String

  • Undocumented

    Declaration

    Swift

    public var tolerance: Float

  • Initializer for Random Forest

    Declaration

    Swift

    public init (data: [[String]],
             target: Int, perform: String,
             using: String, nTrees: Int,
             nFeatures: Int, depth: Int,
             tolerance: Float = 0.1)

    Parameters

    data

    data with labels
    target

    label column number
    perform

    regression or classification
    using

    giniIndex or infoGain
    nTrees

    number of Trees

    Return Value

    Random Forest

  • Makes the forest using bootstrapped data

    Declaration

    Swift

    public func make()

    Return Value

    None

  • Returns bootstrapped and out of bootData

    Declaration

    Swift

    public func splitData(from: [[String]]) -> ([[String]], [[String]])

    Parameters

    from

    string array of data

    Return Value

    boostrapped data and out of boot data

  • Classfies/Predicts an example by using all trees in the foret

    Declaration

    Swift

    public func predict(this: [[String]]) -> String

    Return Value

    classification/predictions as a string

  • Scores the random forest’s accuracy on test data

    Declaration

    Swift

    public func score(with: [[String]]) -> (Float, [String])

    Parameters

    testData

    test data as a 2D string array with feature header

    Return Value

    • accuracy of classifications as float
    • classifications as string array