Expand description
Tree-based algorithms for sklears
This crate provides implementations of tree-based machine learning algorithms including:
- Decision Trees (CART algorithm)
- Random Forest
- Extra Trees
This is a simplified version focusing on core functionality.
Re-exports§
pub use config::ndarray_to_dense_matrix;pub use config::DecisionTreeConfig;pub use config::MaxFeatures;pub use config::MissingValueStrategy;pub use criteria::ConditionalTestType;pub use criteria::FeatureType;pub use criteria::MonotonicConstraint;pub use criteria::SplitCriterion;pub use decision_tree::DecisionTree;pub use decision_tree::DecisionTreeBuilder;pub use decision_tree::DecisionTreeClassifier;pub use decision_tree::DecisionTreeRegressor;pub use decision_tree::TreeValidator;pub use node::CompactTreeNode;pub use node::CustomSplit;pub use node::SurrogateSplit;pub use node::TreeNode;pub use random_forest::RandomForestClassifier;pub use splits::HyperplaneSplit;
Modules§
- builder
- Tree building algorithms and utilities
- config
- Configuration types and enums for decision trees
- criteria
- Split criteria and constraints for decision trees
- decision_
tree - Decision Tree implementation
- node
- Tree node data structures and compact representations
- parallel
- Parallel processing utilities for tree algorithms
- prelude
- Prelude module for convenient imports
- random_
forest - Random Forest implementation using SmartCore
- splits
- Split implementations for decision trees