AdaptiveRandomForestOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for Adaptive Random Forest classifier.

Inheritance: ModelOptions → ClassifierOptions<T> → OnlineClassifierOptions<T> → AdaptiveRandomForestOptions<T>

For Beginners

Adaptive Random Forest (ARF) is an ensemble method that combines multiple Hoeffding trees with drift detection to handle evolving data streams.

Constructors

Constructor Summary
AdaptiveRandomForestOptions()

Properties

Property Summary
double DriftThreshold { get; set; } Gets or sets the drift threshold for tree replacement (DDM).
double HoeffdingDelta { get; set; } Gets or sets the Hoeffding bound confidence parameter for individual trees.
double LambdaPoisson { get; set; } Gets or sets the lambda parameter for Poisson resampling.
double TieThreshold { get; set; } Gets or sets the tie threshold for tree split decisions.
double WarningThreshold { get; set; } Gets or sets the warning threshold for drift detection (DDM).
int GracePeriod { get; set; } Gets or sets the minimum samples before attempting splits.
int MaxTreeDepth { get; set; } Gets or sets the maximum depth for individual trees.
int NumBins { get; set; } Gets or sets the number of bins for numeric attribute discretization.
int NumFeaturesPerTree { get; set; } Gets or sets the number of features to consider for each tree.
int NumTrees { get; set; } Gets or sets the number of trees in the ensemble.