Module learner 

Available on crate feature alloc only.

Unified streaming learner trait for polymorphic model composition.

StreamingLearner is an object-safe trait that abstracts over any online/streaming machine learning model – gradient boosted trees, linear models, Naive Bayes, Mondrian forests, or anything else that can ingest samples one at a time and produce predictions.

Motivation

Stacking ensembles and meta-learners need to treat heterogeneous base models uniformly: train them on the same stream, collect their predictions as features for a combiner, and manage their lifecycle (reset, clone, serialization). StreamingLearner provides exactly this interface.

Object Safety

The trait is deliberately object-safe: every method uses &self / &mut self with concrete return types (no generics on methods, no Self-by-value in non-Sized positions). This means you can store Box<dyn StreamingLearner> in a Vec, enabling runtime-polymorphic stacking without monomorphization.

Capability Traits

Opt-in capability traits narrow the required bound to the actual capability a wrapper or algorithm needs, enabling cleaner type-level documentation:

• HasReadout – models with a linear RLS readout layer (neural models, KRLS, RLS). Used by ProjectedLearner supervised-projection path.
• Tunable – models that expose diagnostics and accept LR / lambda adjustments from AutoML components.
• Structural – models whose capacity can grow or shrink at runtime (tree ensembles: SGBT, ARF, Mondrian).

Traits

HasReadout

Models that expose a linear readout weight vector.

StreamingLearner

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

Structural

Models whose internal capacity can grow or shrink at runtime.

Tunable

Models that expose diagnostics and accept smooth hyperparameter adjustments.