StochTree::GaussianMultivariateRegressionSuffStat Class Reference

Sufficient statistic and associated operations for gaussian homoskedastic constant leaf outcome model. More...

#include <leaf_model.h>

Public Member Functions
	GaussianMultivariateRegressionSuffStat (int basis_dim)
	Construct a new GaussianMultivariateRegressionSuffStat object.
void	IncrementSuffStat (ForestDataset &dataset, Eigen::VectorXd &outcome, ForestTracker &tracker, data_size_t row_idx, int tree_idx)
	Accumulate data from observation row_idx into the sufficient statistics.
void	ResetSuffStat ()
	Reset all of the sufficient statistics to zero.
void	AddSuffStat (GaussianMultivariateRegressionSuffStat &lhs, GaussianMultivariateRegressionSuffStat &rhs)
	Set the value of each sufficient statistic to the sum of the values provided by lhs and rhs
void	SubtractSuffStat (GaussianMultivariateRegressionSuffStat &lhs, GaussianMultivariateRegressionSuffStat &rhs)
	Set the value of each sufficient statistic to the difference between the values provided by lhs and those provided by rhs
bool	SampleGreaterThan (data_size_t threshold)
	Check whether accumulated sample size, n, is greater than some threshold.
bool	SampleGreaterThanEqual (data_size_t threshold)
	Check whether accumulated sample size, n, is greater than or equal to some threshold.
data_size_t	SampleSize ()
	Return the sample size accumulated by a sufficient stat object.

Detailed Description

Sufficient statistic and associated operations for gaussian homoskedastic constant leaf outcome model.

Constructor & Destructor Documentation

GaussianMultivariateRegressionSuffStat()

StochTree::GaussianMultivariateRegressionSuffStat::GaussianMultivariateRegressionSuffStat ( int  basis_dim )

inline

Construct a new GaussianMultivariateRegressionSuffStat object.

Parameters

basis_dim

Size of the basis vector that defines the leaf regression

Member Function Documentation

IncrementSuffStat()

void StochTree::GaussianMultivariateRegressionSuffStat::IncrementSuffStat ( ForestDataset &  dataset, Eigen::VectorXd &  outcome, ForestTracker &  tracker, data_size_t  row_idx, int  tree_idx  )

inline

Accumulate data from observation row_idx into the sufficient statistics.

Parameters

dataset	Data object containining training data, including covariates, leaf regression bases, and case weights
outcome	Data object containing the "partial" residual net of all the model's other mean terms, aside from tree
tracker	Tracking data structures that speed up sampler operations, synchronized with active_forest tracking a forest's state
row_idx	Index of the training data observation from which the sufficient statistics should be updated
tree_idx	Index of the tree being updated in the course of this sufficient statistic update

AddSuffStat()

void StochTree::GaussianMultivariateRegressionSuffStat::AddSuffStat ( GaussianMultivariateRegressionSuffStat &  lhs, GaussianMultivariateRegressionSuffStat &  rhs  )

inline

Set the value of each sufficient statistic to the sum of the values provided by lhs and rhs

Parameters

lhs	First sufficient statistic ("left hand side")
rhs	Second sufficient statistic ("right hand side")

SubtractSuffStat()

void StochTree::GaussianMultivariateRegressionSuffStat::SubtractSuffStat ( GaussianMultivariateRegressionSuffStat &  lhs, GaussianMultivariateRegressionSuffStat &  rhs  )

inline

Set the value of each sufficient statistic to the difference between the values provided by lhs and those provided by rhs

Parameters

lhs	First sufficient statistic ("left hand side")
rhs	Second sufficient statistic ("right hand side")

SampleGreaterThan()

bool StochTree::GaussianMultivariateRegressionSuffStat::SampleGreaterThan ( data_size_t  threshold )

inline

Check whether accumulated sample size, n, is greater than some threshold.

Parameters

threshold

Value used to compute n > threshold

SampleGreaterThanEqual()

bool StochTree::GaussianMultivariateRegressionSuffStat::SampleGreaterThanEqual ( data_size_t  threshold )

inline

Check whether accumulated sample size, n, is greater than or equal to some threshold.

Parameters

threshold

Value used to compute n >= threshold

---

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

• include/stochtree/leaf_model.h