ExtraTreesClassifierOptions<T>
Options & Configuration · AiDotNet.Models.Options
Configuration options for Extra Trees (Extremely Randomized Trees) classifier.
Inheritance: ModelOptions → ClassifierOptions<T> → ExtraTreesClassifierOptions<T>
For Beginners
Extra Trees is a “more random” version of Random Forest!
The key differences from Random Forest:
- Random Forest finds the BEST split among random features
- Extra Trees picks RANDOM splits for random features
This extra randomness:
- Makes training even faster
- Often generalizes better (less overfitting)
- Creates more diverse trees
When to use Extra Trees:
- When Random Forest is overfitting
- When you need faster training
- As an alternative to try alongside Random Forest
How It Works
Extra Trees is similar to Random Forest but with even more randomization. Instead of finding the best split, it selects splits at random, which can lead to better generalization and faster training.
Constructors
| Constructor | Summary |
|---|---|
ExtraTreesClassifierOptions() |
Properties
| Property | Summary |
|---|---|
ClassificationSplitCriterion Criterion { get; set; } |
Gets or sets the split criterion. |
MaxFeatureSelection MaxFeatures { get; set; } |
Gets or sets the rule for how many features each split considers. |
bool Bootstrap { get; set; } |
Gets or sets whether to bootstrap samples. |
double MinImpurityDecrease { get; set; } |
Gets or sets the minimum impurity decrease for splitting. |
int MinSamplesLeaf { get; set; } |
Gets or sets the minimum samples required at a leaf. |
int MinSamplesSplit { get; set; } |
Gets or sets the minimum samples required to split. |
int NEstimators { get; set; } |
Gets or sets the number of trees in the forest. |
int? MaxDepth { get; set; } |
Gets or sets the maximum depth of each tree. |
int? MaxFeatureCount { get; set; } |
Gets or sets an explicit feature count per split, overriding MaxFeatures. |