ppforest2::ClassificationForest Class Reference

Random forest of classification trees. More...

#include <ClassificationForest.hpp>

Public Types
using	FeatureMatrix = types::FeatureMatrix
using	FeatureVector = types::FeatureVector
using	GroupIndices = std::map<types::GroupId, int>
using	GroupPartition = stats::GroupPartition
using	Groups = std::set<types::GroupId>
using	Outcome = types::Outcome
using	OutcomeVector = types::OutcomeVector
using	Ptr = std::unique_ptr<ClassificationForest>
using	RNG = stats::RNG
Public Types inherited from ppforest2::Forest
using	FeatureMatrix = types::FeatureMatrix
using	FeatureVector = types::FeatureVector
using	Outcome = types::Outcome
using	OutcomeVector = types::OutcomeVector
using	Ptr = std::unique_ptr<Forest>
using	RNG = stats::RNG
Public Types inherited from ppforest2::Model
using	Ptr = std::shared_ptr<Model>

Public Member Functions
	ClassificationForest (TrainingSpec::Ptr spec, Groups groups)
void	accept (Model::Visitor &visitor) const override
	Accept a model visitor (mode-specific dispatch).
FeatureMatrix	predict (FeatureMatrix const &x, Proportions) const
	Vote proportions per row, columns laid out by groups.
FeatureMatrix	predict (FeatureMatrix const &x, Proportions, GroupIndices const &indices) const
	Vote proportions per row, with an explicit column layout.
types::Outcome	predict (FeatureVector const &x) const override
	Per-row prediction (mode-specific: majority vote or mean).
FeatureVector	predict (FeatureVector const &x, Proportions) const
	Vote proportions for one observation, columns laid out by groups.
FeatureVector	predict (FeatureVector const &x, Proportions, GroupIndices const &indices) const
	Vote proportions for one observation, with an explicit column layout passed as a precomputed {group → column} map.
Public Member Functions inherited from ppforest2::Forest
void	add_tree (BaggedTree::Ptr tree)
	Add a trained bagged tree to the forest.
bool	operator!= (Forest const &other) const
bool	operator== (Forest const &other) const
virtual types::OutcomeVector	predict (types::FeatureMatrix const &x) const
	Predict a matrix of observations.
Public Member Functions inherited from ppforest2::Model
virtual	~Model ()=default

Static Public Member Functions
static Ptr	train (TrainingSpec const &spec, FeatureMatrix const &x, OutcomeVector const &y)
Static Public Member Functions inherited from ppforest2::Forest
static Ptr	train (TrainingSpec const &spec, FeatureMatrix const &x, OutcomeVector const &y)
	Train a random forest.
Static Public Member Functions inherited from ppforest2::Model
static void	check_train_inputs (types::FeatureMatrix const &x, types::OutcomeVector const &y)
	Validate common training inputs (y non-empty, matching x rows).
static Ptr	train (TrainingSpec const &spec, types::FeatureMatrix &x, types::OutcomeVector &y)
	Train a model from a training specification.

Public Attributes
Groups	groups
	Set of group labels this forest predicts over.
Public Attributes inherited from ppforest2::Forest
std::vector< BaggedTree::Ptr >	trees
	Bootstrap-aggregated trees. Each BaggedTree pairs the polymorphic inner Tree with its sample indices.
Public Attributes inherited from ppforest2::Model
bool	degenerate = false
	Whether the model contains degenerate nodes/splits.
TrainingSpec::Ptr	training_spec
	Training specification used to build this model.

Protected Member Functions
BaggedTree::Ptr	train_tree (FeatureMatrix const &x, OutcomeVector const &y, RNG &rng) const override
	Train one bagged tree on a bootstrap resample of x / y.
Protected Member Functions inherited from ppforest2::Forest
	Forest ()=default
	Forest (TrainingSpec::Ptr spec)
void	build_trees (types::FeatureMatrix const &x, types::OutcomeVector const &y)
	Build training_spec->size bagged trees in parallel and attach them to this forest.

Detailed Description

Random forest of classification trees.

Aggregates per-tree predictions by majority vote. Use the free functions in Metrics.hpp for OOB diagnostics and variable importance, and predict_proportions (in Model.hpp) for per-group vote proportions.

Member Typedef Documentation

FeatureMatrix

using ppforest2::ClassificationForest::FeatureMatrix = types::FeatureMatrix

FeatureVector

using ppforest2::ClassificationForest::FeatureVector = types::FeatureVector

GroupIndices

using ppforest2::ClassificationForest::GroupIndices = std::map<types::GroupId, int>

GroupPartition

using ppforest2::ClassificationForest::GroupPartition = stats::GroupPartition

Groups

using ppforest2::ClassificationForest::Groups = std::set<types::GroupId>

Outcome

using ppforest2::ClassificationForest::Outcome = types::Outcome

OutcomeVector

using ppforest2::ClassificationForest::OutcomeVector = types::OutcomeVector

Ptr

using ppforest2::ClassificationForest::Ptr = std::unique_ptr<ClassificationForest>

RNG

using ppforest2::ClassificationForest::RNG = stats::RNG

Constructor & Destructor Documentation

ClassificationForest()

ppforest2::ClassificationForest::ClassificationForest ( TrainingSpec::Ptr spec, Groups groups )

inline

Member Function Documentation

accept()

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

overridevirtual

Accept a model visitor (mode-specific dispatch).

Implements ppforest2::Forest.

predict() [1/5]

FeatureMatrix ppforest2::ClassificationForest::predict	(	FeatureMatrix const &	x,
		Proportions	) const

Vote proportions per row, columns laid out by groups.

predict() [2/5]

FeatureMatrix ppforest2::ClassificationForest::predict	(	FeatureMatrix const &	x,
		Proportions	,
		GroupIndices const &	indices ) const

Vote proportions per row, with an explicit column layout.

Iterates rows and calls the single-row predict(..., indices) overload — indices is built zero times here.

predict() [3/5]

types::Outcome ppforest2::ClassificationForest::predict ( FeatureVector const & x ) const

overridevirtual

Per-row prediction (mode-specific: majority vote or mean).

Implements ppforest2::Forest.

predict() [4/5]

FeatureVector ppforest2::ClassificationForest::predict	(	FeatureVector const &	x,
		Proportions	) const

Vote proportions for one observation, columns laid out by groups.

predict() [5/5]

FeatureVector ppforest2::ClassificationForest::predict	(	FeatureVector const &	x,
		Proportions	,
		GroupIndices const &	indices ) const

Vote proportions for one observation, with an explicit column layout passed as a precomputed {group → column} map.

Base primitive. The matrix overload calls this once per row with the same indices, so the map is built only once per predict(matrix, ...) call rather than n times.

train()

static Ptr ppforest2::ClassificationForest::train ( TrainingSpec const & spec, FeatureMatrix const & x, OutcomeVector const & y )

static

train_tree()

BaggedTree::Ptr ppforest2::ClassificationForest::train_tree ( FeatureMatrix const & x, OutcomeVector const & y, RNG & rng ) const

overrideprotectedvirtual

Train one bagged tree on a bootstrap resample of x / y.

Mode-specific hook invoked by build_trees once per slot in the forest (with retries on degenerate trees). Subclasses that need per-training shared state (e.g. ClassificationForest caches the parent's label partition for stratified sampling) hold it as a private transient pointer set up before build_trees runs.

Called from inside build_trees's OpenMP parallel for, so implementations must be thread-safe and read-only over *this.

Implements ppforest2::Forest.

Member Data Documentation

groups

Groups ppforest2::ClassificationForest::groups

Set of group labels this forest predicts over.

Canonical column layout for predict(FeatureMatrix, Proportions). Populated at construction (training: from y_part.groups; JSON: from meta.groups indices; R: from forest$groups names). Independent of any individual tree's node_groups() — a degenerate-retry tree with a narrower group set can't shrink the prediction matrix.

---

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

• core/src/models/ClassificationForest.hpp