Crate ferrolearn_tree 

ferrolearn-tree

Decision tree and ensemble tree models for the ferrolearn machine learning framework.

This crate provides implementations of:

• DecisionTreeClassifier / DecisionTreeRegressor — CART decision trees with configurable splitting criteria, depth limits, and minimum sample constraints.
• RandomForestClassifier / RandomForestRegressor — Bootstrap-aggregated ensembles of decision trees with random feature subsets, built in parallel via rayon.
• GradientBoostingClassifier / GradientBoostingRegressor — Gradient boosting ensembles that sequentially fit trees to the negative gradient of a loss function.
• HistGradientBoostingClassifier / HistGradientBoostingRegressor — Histogram-based gradient boosting with O(n_bins) split finding, subtraction trick, native NaN support, and optional best-first (leaf-wise) growth.
• AdaBoostClassifier — Adaptive Boosting using decision tree stumps with SAMME and SAMME.R algorithms.
• ExtraTreeClassifier / ExtraTreeRegressor — Extremely randomized trees where split thresholds are chosen randomly rather than via exhaustive search.
• ExtraTreesClassifier / ExtraTreesRegressor — Ensembles of extremely randomized trees with Rayon parallel fitting. No bootstrap by default.
• IsolationForest — Anomaly detection via random isolation trees.
• VotingClassifier / VotingRegressor — Ensembles of decision trees with varying hyperparameters, aggregated by majority vote or averaging.
• RandomTreesEmbedding — Unsupervised feature transformation via one-hot encoded leaf indices across an ensemble of randomly built trees.

Design

Each model follows the compile-time safety pattern:

• The unfitted struct (e.g., DecisionTreeClassifier<F>) holds hyperparameters and implements Fit.
• Calling fit() produces a new fitted type (e.g., FittedDecisionTreeClassifier<F>) that implements Predict.
• Calling predict() on an unfitted model is a compile-time error.

Pipeline Integration

All models implement PipelineEstimator for f64, allowing them to be used as the final step in a Pipeline.

Float Generics

All models are generic over F: num_traits::Float + Send + Sync + 'static, supporting both f32 and f64.

Re-exports

pub use adaboost::AdaBoostAlgorithm;

pub use adaboost::AdaBoostClassifier;

pub use adaboost::FittedAdaBoostClassifier;

pub use decision_tree::ClassificationCriterion;

pub use decision_tree::DecisionTreeClassifier;

pub use decision_tree::DecisionTreeRegressor;

pub use decision_tree::FittedDecisionTreeClassifier;

pub use decision_tree::FittedDecisionTreeRegressor;

pub use decision_tree::Node;

pub use decision_tree::RegressionCriterion;

pub use extra_tree::ExtraTreeClassifier;

pub use extra_tree::ExtraTreeRegressor;

pub use extra_tree::FittedExtraTreeClassifier;

pub use extra_tree::FittedExtraTreeRegressor;

pub use extra_trees_ensemble::ExtraTreesClassifier;

pub use extra_trees_ensemble::ExtraTreesRegressor;

pub use extra_trees_ensemble::FittedExtraTreesClassifier;

pub use extra_trees_ensemble::FittedExtraTreesRegressor;

pub use gradient_boosting::ClassificationLoss;

pub use gradient_boosting::FittedGradientBoostingClassifier;

pub use gradient_boosting::FittedGradientBoostingRegressor;

pub use gradient_boosting::GradientBoostingClassifier;

pub use gradient_boosting::GradientBoostingRegressor;

pub use gradient_boosting::RegressionLoss;

pub use hist_gradient_boosting::FittedHistGradientBoostingClassifier;

pub use hist_gradient_boosting::FittedHistGradientBoostingRegressor;

pub use hist_gradient_boosting::HistClassificationLoss;

pub use hist_gradient_boosting::HistGradientBoostingClassifier;

pub use hist_gradient_boosting::HistGradientBoostingRegressor;

pub use hist_gradient_boosting::HistNode;

pub use hist_gradient_boosting::HistRegressionLoss;

pub use isolation_forest::FittedIsolationForest;

pub use isolation_forest::IsolationForest;

pub use random_forest::FittedRandomForestClassifier;

pub use random_forest::FittedRandomForestRegressor;

pub use random_forest::MaxFeatures;

pub use random_forest::RandomForestClassifier;

pub use random_forest::RandomForestRegressor;

pub use random_trees_embedding::FittedRandomTreesEmbedding;

pub use random_trees_embedding::RandomTreesEmbedding;

pub use voting::FittedVotingClassifier;

pub use voting::FittedVotingRegressor;

pub use voting::VotingClassifier;

pub use voting::VotingRegressor;

Modules

adaboost

AdaBoost classifier.

decision_tree

CART decision tree classifiers and regressors.

extra_tree

Extremely randomized tree classifiers and regressors.

extra_trees_ensemble

Extremely randomized trees ensemble classifiers and regressors.

gradient_boosting

Gradient boosting classifiers and regressors.

hist_gradient_boosting

Histogram-based gradient boosting classifiers and regressors.

isolation_forest

Isolation forest anomaly detection.

random_forest

Random forest classifiers and regressors.

random_trees_embedding

Random trees embedding for unsupervised feature transformation.

voting

Voting ensemble classifiers and regressors.