ppforest2::Model Struct Reference

Abstract base class for predictive models (trees and forests). More...

#include <Model.hpp>

Classes
struct	Visitor
	Visitor interface for model dispatch. More...

Public Types
using	Ptr = std::shared_ptr<Model>

Public Member Functions
virtual	~Model ()=default
virtual void	accept (Visitor &visitor) const =0
	Accept a model visitor (double dispatch).
virtual types::ResponseVector	predict (types::FeatureMatrix const &data) const =0
	Predict a matrix of observations.
virtual types::FeatureMatrix	predict (types::FeatureMatrix const &data, Proportions) const =0
	Predict proportions for a matrix of observations.
virtual types::Response	predict (types::FeatureVector const &data) const =0
	Predict a single observation.

Static Public Member Functions
static Ptr	train (TrainingSpec const &spec, types::FeatureMatrix const &x, types::ResponseVector const &y)
	Train a model from a training specification.

Public Attributes
bool	degenerate = false
	Whether the model contains degenerate nodes/splits.
TrainingSpec::Ptr	training_spec
	Training specification used to build this model.

Detailed Description

Abstract base class for predictive models (trees and forests).

Member Typedef Documentation

Ptr

using ppforest2::Model::Ptr = std::shared_ptr<Model>

Constructor & Destructor Documentation

~Model()

virtual ppforest2::Model::~Model ( )

virtualdefault

Member Function Documentation

accept()

virtual void ppforest2::Model::accept ( Visitor & visitor ) const

pure virtual

Accept a model visitor (double dispatch).

Implemented in ppforest2::Forest, and ppforest2::Tree.

predict() [1/3]

virtual types::ResponseVector ppforest2::Model::predict ( types::FeatureMatrix const & data ) const

pure virtual

Predict a matrix of observations.

Parameters

data	Feature matrix (n × p).

Returns

Predicted group labels (n).

Implemented in ppforest2::Forest, and ppforest2::Tree.

predict() [2/3]

virtual types::FeatureMatrix ppforest2::Model::predict ( types::FeatureMatrix const & data, Proportions  ) const

pure virtual

Predict proportions for a matrix of observations.

Returns an (n × G) matrix. For forests, each row contains the fraction of trees that voted for each group. For single trees, each row is a one-hot encoding of the predicted group.

Parameters

data	Feature matrix (n × p).

Returns

Proportion matrix (n × G), rows sum to 1.0.

Implemented in ppforest2::Forest, and ppforest2::Tree.

predict() [3/3]

virtual types::Response ppforest2::Model::predict ( types::FeatureVector const & data ) const

pure virtual

Predict a single observation.

Parameters

data	Feature vector (p).

Returns

Predicted group label.

Implemented in ppforest2::Forest, and ppforest2::Tree.

train()

static Ptr ppforest2::Model::train ( TrainingSpec const & spec, types::FeatureMatrix const & x, types::ResponseVector const & y )

static

Train a model from a training specification.

Dispatches to Tree::train or Forest::train based on spec.is_forest().

Parameters

spec	Training specification.
x	Feature matrix (n × p).
y	Response vector (n).

Returns

Trained model (Tree or Forest).

Member Data Documentation

degenerate

bool ppforest2::Model::degenerate = false

Whether the model contains degenerate nodes/splits.

training_spec

TrainingSpec::Ptr ppforest2::Model::training_spec

Training specification used to build this model.

---

The documentation for this struct was generated from the following file:

• core/src/models/Model.hpp