Trait StreamingLearner 

pub trait StreamingLearner: Send + Sync {
    // Required methods
    fn train_one(&mut self, features: &[f64], target: f64, weight: f64);
    fn predict(&self, features: &[f64]) -> f64;
    fn n_samples_seen(&self) -> u64;
    fn reset(&mut self);

    // Provided methods
    fn train(&mut self, features: &[f64], target: f64) { ... }
    fn predict_batch(&self, feature_matrix: &[&[f64]]) -> Vec<f64> { ... }
    fn diagnostics_array(&self) -> [f64; 5] { ... }
    fn adjust_config(&mut self, _lr_multiplier: f64, _lambda_delta: f64) { ... }
    fn apply_structural_change(&mut self, _depth_delta: i32, _steps_delta: i32) { ... }
    fn replacement_count(&self) -> u64 { ... }
}

Object-safe trait for any streaming (online) machine learning model.

All methods use &self or &mut self with concrete return types, ensuring the trait can be used behind Box<dyn StreamingLearner> for runtime-polymorphic stacking ensembles.

The Send + Sync supertraits allow learners to be shared across threads (e.g., for parallel prediction in async pipelines).

Required Methods

Method	Purpose
train_one	Ingest a single weighted observation
predict	Produce a prediction for a feature vector
n_samples_seen	Total observations ingested so far
reset	Clear all learned state, returning to a fresh model

Default Methods

Method	Purpose
train	Convenience wrapper calling train_one with unit weight
predict_batch	Map predict over a slice of feature vectors
diagnostics_array	Raw diagnostic signals for adaptive tuning (all zeros by default)
adjust_config	Apply smooth LR/lambda adjustments (no-op by default)
apply_structural_change	Apply depth/steps changes at replacement boundaries (no-op by default)
replacement_count	Total internal model replacements (0 by default)

Required Methods

fn train_one(&mut self, features: &[f64], target: f64, weight: f64)

Train on a single observation with explicit sample weight.

This is the fundamental training primitive. All streaming models must support weighted incremental updates – even if the weight is simply used to scale gradient contributions.

Arguments

• features – feature vector for this observation
• target – target value (regression) or class label (classification)
• weight – sample weight (1.0 for uniform weighting)

fn predict(&self, features: &[f64]) -> f64

Predict the target for the given feature vector.

Returns the raw model output (no loss transform applied). For SGBT this is the sum of tree predictions; for linear models this is the dot product plus bias.

fn n_samples_seen(&self) -> u64

Total number of observations trained on since creation or last reset.

fn reset(&mut self)

Reset the model to its initial (untrained) state.

After calling reset(), the model should behave identically to a freshly constructed instance with the same configuration. In particular, n_samples_seen() must return 0.

Provided Methods

fn train(&mut self, features: &[f64], target: f64)

Train on a single observation with unit weight.

Convenience wrapper around train_one that passes weight = 1.0. This is the most common training call in practice.

fn predict_batch(&self, feature_matrix: &[&[f64]]) -> Vec<f64>

Predict for each row in a feature matrix.

Returns a Vec<f64> with one prediction per input row. The default implementation simply maps predict over the slices; concrete implementations may override this for SIMD or batch-optimized prediction paths.

Arguments

• feature_matrix – each element is a feature vector (one row)

fn diagnostics_array(&self) -> [f64; 5]

Raw diagnostic signals for adaptive tuning.

Returns [residual_alignment, reg_sensitivity, depth_sufficiency, effective_dof, uncertainty]. These five signals drive the diagnostic adaptor in the auto-builder pipeline.

Default: all zeros (model does not provide diagnostics). Models with internal diagnostic caches (e.g. SGBT, DistributionalSGBT) override this to return real computed values.

fn adjust_config(&mut self, _lr_multiplier: f64, _lambda_delta: f64)

Apply smooth learning rate and regularization adjustments.

• lr_multiplier – scales the current learning rate (1.0 = no change, 0.99 = 1% decrease, 1.01 = 1% increase).
• lambda_delta – added to the L2 regularization parameter (0.0 = no change, positive = increase, negative = decrease).

Default: no-op. Override for models with adjustable hyperparameters (e.g. SGBT, DistributionalSGBT).

fn apply_structural_change(&mut self, _depth_delta: i32, _steps_delta: i32)

Apply structural changes at model replacement boundaries.

• depth_delta – adjust maximum tree depth (+1, -1, or 0).
• steps_delta – adjust number of ensemble steps (+2, -2, or 0).

Structural changes take effect on the next tree replacement, not immediately. Default: no-op for models without structural config.

fn replacement_count(&self) -> u64

Total number of internal model replacements (e.g. tree replacements triggered by drift detection or max-tree-samples).

External callers (e.g. the auto-builder) use this to detect when a structural boundary has occurred and apply queued structural changes. Default: 0 for models without replacement semantics.

Implementors

impl StreamingLearner for StreamingAttentionModel

impl StreamingLearner for AutoTuner

impl StreamingLearner for ContinualLearner

impl StreamingLearner for irithyll::ensemble::adaptive_forest::AdaptiveRandomForest

impl StreamingLearner for irithyll::ensemble::distributional::DistributionalSGBT

impl StreamingLearner for irithyll::ensemble::moe_distributional::MoEDistributionalSGBT

impl StreamingLearner for irithyll::ensemble::stacked::StackedEnsemble

impl StreamingLearner for irithyll_core::ensemble::adaptive_forest::AdaptiveRandomForest

impl StreamingLearner for irithyll_core::ensemble::distributional::DistributionalSGBT

impl StreamingLearner for irithyll_core::ensemble::moe_distributional::MoEDistributionalSGBT

impl StreamingLearner for irithyll_core::ensemble::stacked::StackedEnsemble

impl StreamingLearner for StreamingKAN

impl StreamingLearner for ClassificationWrapper

impl StreamingLearner for KRLS

impl StreamingLearner for StreamingLinearModel

impl StreamingLearner for MondrianForest

impl StreamingLearner for BernoulliNB

impl StreamingLearner for MultinomialNB

impl StreamingLearner for GaussianNB

impl StreamingLearner for LocallyWeightedRegression

impl StreamingLearner for RecursiveLeastSquares

impl StreamingLearner for StreamingPolynomialRegression

impl StreamingLearner for NeuralMoE

impl StreamingLearner for Pipeline

impl StreamingLearner for EchoStateNetwork

impl StreamingLearner for NextGenRC

impl StreamingLearner for SpikeNet

impl StreamingLearner for StreamingMamba

impl StreamingLearner for HoeffdingTreeClassifier

impl StreamingLearner for HoltWinters

impl StreamingLearner for SNARIMAX

StreamingLearner implementation for SNARIMAX.

The features parameter maps to exogenous inputs, and target maps to the observed time series value. Sample weight is accepted but currently unused (all observations are weighted equally in the SGD update).

impl StreamingLearner for StreamingTTT

impl<L> StreamingLearner for irithyll_core::ensemble::adaptive::AdaptiveSGBT<L>

where L: Loss,

impl<L> StreamingLearner for irithyll_core::ensemble::bagged::BaggedSGBT<L>

where L: Loss + Clone,

impl<L> StreamingLearner for irithyll_core::ensemble::moe::MoESGBT<L>

where L: Loss,

impl<L: Loss + Clone> StreamingLearner for irithyll::ensemble::bagged::BaggedSGBT<L>

impl<L: Loss> StreamingLearner for irithyll::ensemble::adaptive::AdaptiveSGBT<L>

impl<L: Loss> StreamingLearner for irithyll::ensemble::moe::MoESGBT<L>

impl<L: Loss> StreamingLearner for SGBTLearner<L>