Enum Model 

pub enum Model {
    DecisionTreeClassifier(DecisionTreeClassifier),
    DecisionTreeRegressor(DecisionTreeRegressor),
    RandomForest(RandomForest),
    GradientBoostedTrees(GradientBoostedTrees),
}

Top-level semantic model enum.

This type stays close to the learned structure rather than the fastest possible runtime layout. That is what makes it suitable for introspection, serialization, and exact behavior parity across bindings.

Variants

DecisionTreeClassifier(DecisionTreeClassifier)

DecisionTreeRegressor(DecisionTreeRegressor)

RandomForest(RandomForest)

GradientBoostedTrees(GradientBoostedTrees)

Implementations

impl Model

pub fn predict_table(&self, table: &dyn TableAccess) -> Vec<f64>

pub fn predict_rows( &self, rows: Vec<Vec<f64>>, ) -> Result<Vec<f64>, PredictError>

pub fn predict_proba_table( &self, table: &dyn TableAccess, ) -> Result<Vec<Vec<f64>>, PredictError>

pub fn predict_proba_rows( &self, rows: Vec<Vec<f64>>, ) -> Result<Vec<Vec<f64>>, PredictError>

pub fn predict_named_columns( &self, columns: BTreeMap<String, Vec<f64>>, ) -> Result<Vec<f64>, PredictError>

pub fn predict_proba_named_columns( &self, columns: BTreeMap<String, Vec<f64>>, ) -> Result<Vec<Vec<f64>>, PredictError>

pub fn predict_sparse_binary_columns( &self, n_rows: usize, n_features: usize, columns: Vec<Vec<usize>>, ) -> Result<Vec<f64>, PredictError>

pub fn predict_proba_sparse_binary_columns( &self, n_rows: usize, n_features: usize, columns: Vec<Vec<usize>>, ) -> Result<Vec<Vec<f64>>, PredictError>

pub fn algorithm(&self) -> TrainAlgorithm

pub fn task(&self) -> Task

pub fn criterion(&self) -> Criterion

pub fn tree_type(&self) -> TreeType

pub fn mean_value(&self) -> Option<f64>

pub fn canaries(&self) -> usize

pub fn max_depth(&self) -> Option<usize>

pub fn min_samples_split(&self) -> Option<usize>

pub fn min_samples_leaf(&self) -> Option<usize>

pub fn n_trees(&self) -> Option<usize>

pub fn max_features(&self) -> Option<usize>

pub fn seed(&self) -> Option<u64>

pub fn compute_oob(&self) -> bool

pub fn oob_score(&self) -> Option<f64>

pub fn learning_rate(&self) -> Option<f64>

pub fn bootstrap(&self) -> bool

pub fn top_gradient_fraction(&self) -> Option<f64>

pub fn other_gradient_fraction(&self) -> Option<f64>

pub fn tree_count(&self) -> usize

pub fn tree_structure( &self, tree_index: usize, ) -> Result<TreeStructureSummary, IntrospectionError>

pub fn tree_prediction_stats( &self, tree_index: usize, ) -> Result<PredictionValueStats, IntrospectionError>

pub fn tree_node( &self, tree_index: usize, node_index: usize, ) -> Result<NodeTreeNode, IntrospectionError>

pub fn tree_level( &self, tree_index: usize, level_index: usize, ) -> Result<ObliviousLevel, IntrospectionError>

pub fn tree_leaf( &self, tree_index: usize, leaf_index: usize, ) -> Result<IndexedLeaf, IntrospectionError>

pub fn to_ir(&self) -> ModelPackageIr

pub fn to_ir_json(&self) -> Result<String, Error>

pub fn to_ir_json_pretty(&self) -> Result<String, Error>

pub fn serialize(&self) -> Result<String, Error>

pub fn serialize_pretty(&self) -> Result<String, Error>

pub fn optimize_inference( &self, physical_cores: Option<usize>, ) -> Result<OptimizedModel, OptimizeError>

pub fn json_schema() -> RootSchema

pub fn json_schema_json() -> Result<String, IrError>

pub fn json_schema_json_pretty() -> Result<String, IrError>

pub fn deserialize(serialized: &str) -> Result<Self, IrError>

pub fn used_feature_indices(&self) -> Vec<usize>

pub fn used_feature_count(&self) -> usize

Trait Implementations

impl Clone for Model

fn clone(&self) -> Model

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Model

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.