ITreeBasedRegression<T>
Interfaces · AiDotNet.Interfaces
Defines the core functionality for tree-based machine learning models.
Implements: INonLinearRegression<T>
Known implementations: DecisionTreeRegressionBase<T>, IAsyncTreeBasedModel<T>
How It Works
Tree-based models make predictions by following a series of decision rules organized in a tree-like structure. These models can be used for both classification (predicting categories) and regression (predicting numeric values).
For Beginners: Tree-based models work like a flowchart of yes/no questions to make predictions. Imagine you’re trying to predict if someone will like a movie:
- Is it an action movie? If yes, go to question 2. If no, go to question 3.
- Does it have their favorite actor? If yes, predict “Like”. If no, predict “Dislike”.
- Is it less than 2 hours long? If yes, predict “Like”. If no, predict “Dislike”.
This is a simple decision tree. More advanced tree-based models like Random Forests or Gradient Boosted Trees use multiple trees together to make better predictions.
This interface inherits from IFullModel
Properties
| Property | Summary |
|---|---|
Vector<T> FeatureImportances { get; } |
Gets the relative importance of each feature in making predictions. |
int MaxDepth { get; } |
Gets the maximum depth (number of sequential decisions) allowed in each decision tree. |
int NumberOfTrees { get; } |
Gets the number of decision trees used in the model. |