ForestSamples: Class that stores draws from an random ensemble of decision...
In stochtree: Stochastic Tree Ensembles (XBART and BART) for Supervised Learning and Causal Inference
ForestSamples R Documentation
Class that stores draws from an random ensemble of decision trees
Description
Wrapper around a C++ container of tree ensembles
Public fields
forest_container_ptrExternal pointer to a C++ ForestContainer class
Methods
Public methods
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Method new()
Create a new ForestContainer object.
Usage
ForestSamples$new(
num_trees,
leaf_dimension = 1,
is_leaf_constant = FALSE,
is_exponentiated = FALSE
)
Arguments
num_treesNumber of trees
leaf_dimensionDimensionality of the outcome model
is_leaf_constantWhether leaf is constant
is_exponentiatedWhether forest predictions should be exponentiated before being returned
Returns
A new ForestContainer object.
Method load_from_json()
Create a new ForestContainer object from a json object
Usage
ForestSamples$load_from_json(json_object, json_forest_label)
Arguments
json_objectObject of class CppJson
json_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
A new ForestContainer object.
Method append_from_json()
Append to a ForestContainer object from a json object
Usage
ForestSamples$append_from_json(json_object, json_forest_label)
Arguments
json_objectObject of class CppJson
json_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
None
Method load_from_json_string()
Create a new ForestContainer object from a json object
Usage
ForestSamples$load_from_json_string(json_string, json_forest_label)
Arguments
json_stringJSON string which parses into object of class CppJson
json_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
A new ForestContainer object.
Method append_from_json_string()
Append to a ForestContainer object from a json object
Usage
ForestSamples$append_from_json_string(json_string, json_forest_label)
Arguments
json_stringJSON string which parses into object of class CppJson
json_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
None
Method predict()
Predict every tree ensemble on every sample in forest_dataset
Usage
ForestSamples$predict(forest_dataset)
Arguments
forest_datasetForestDataset R class
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as samples in the ForestContainer
Method predict_raw()
Predict "raw" leaf values (without being multiplied by basis) for every tree ensemble on every sample in forest_dataset
Usage
ForestSamples$predict_raw(forest_dataset)
Arguments
forest_datasetForestDataset R class
Returns
Array of predictions for each observation in forest_dataset and
each sample in the ForestSamples class with each prediction having the
dimensionality of the forests' leaf model. In the case of a constant leaf model
or univariate leaf regression, this array is two-dimensional (number of observations,
number of forest samples). In the case of a multivariate leaf regression,
this array is three-dimension (number of observations, leaf model dimension,
number of samples).
Method predict_raw_single_forest()
Predict "raw" leaf values (without being multiplied by basis) for a specific forest on every sample in forest_dataset
Usage
ForestSamples$predict_raw_single_forest(forest_dataset, forest_num)
Arguments
forest_datasetForestDataset R class
forest_numIndex of the forest sample within the container
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as dimensions in the leaves of trees in ForestContainer
Method predict_raw_single_tree()
Predict "raw" leaf values (without being multiplied by basis) for a specific tree in a specific forest on every observation in forest_dataset
Usage
ForestSamples$predict_raw_single_tree(forest_dataset, forest_num, tree_num)
Arguments
forest_datasetForestDataset R class
forest_numIndex of the forest sample within the container
tree_numIndex of the tree to be queried
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as dimensions in the leaves of trees in ForestContainer
Method set_root_leaves()
Set a constant predicted value for every tree in the ensemble.
Stops program if any tree is more than a root node.
Usage
ForestSamples$set_root_leaves(forest_num, leaf_value)
Arguments
forest_numIndex of the forest sample within the container.
leaf_valueConstant leaf value(s) to be fixed for each tree in the ensemble indexed by forest_num. Can be either a single number or a vector, depending on the forest's leaf dimension.
Method prepare_for_sampler()
Set a constant predicted value for every tree in the ensemble.
Stops program if any tree is more than a root node.
Usage
ForestSamples$prepare_for_sampler(
dataset,
outcome,
forest_model,
leaf_model_int,
leaf_value
)
Arguments
datasetForestDataset Dataset class (covariates, basis, etc...)
outcomeOutcome Outcome class (residual / partial residual)
forest_modelForestModel object storing tracking structures used in training / sampling
leaf_model_intInteger value encoding the leaf model type (0 = constant gaussian, 1 = univariate gaussian, 2 = multivariate gaussian, 3 = log linear variance).
leaf_valueConstant leaf value(s) to be fixed for each tree in the ensemble indexed by forest_num. Can be either a single number or a vector, depending on the forest's leaf dimension.
Method adjust_residual()
Adjusts residual based on the predictions of a forest
This is typically run just once at the beginning of a forest sampling algorithm.
After trees are initialized with constant root node predictions, their root predictions are subtracted out of the residual.
Usage
ForestSamples$adjust_residual(
dataset,
outcome,
forest_model,
requires_basis,
forest_num,
add
)
Arguments
datasetForestDataset object storing the covariates and bases for a given forest
outcomeOutcome object storing the residuals to be updated based on forest predictions
forest_modelForestModel object storing tracking structures used in training / sampling
requires_basisWhether or not a forest requires a basis for prediction
forest_numIndex of forest used to update residuals
addWhether forest predictions should be added to or subtracted from residuals
Method save_json()
Store the trees and metadata of ForestDataset class in a json file
Usage
ForestSamples$save_json(json_filename)
Arguments
json_filenameName of output json file (must end in ".json")
Method load_json()
Load trees and metadata for an ensemble from a json file. Note that
any trees and metadata already present in ForestDataset class will
be overwritten.
Usage
ForestSamples$load_json(json_filename)
Arguments
json_filenameName of model input json file (must end in ".json")
Method num_samples()
Return number of samples in a ForestContainer object
Usage
ForestSamples$num_samples()
Returns
Sample count
Method num_trees()
Return number of trees in each ensemble of a ForestContainer object
Usage
ForestSamples$num_trees()
Returns
Tree count
Method leaf_dimension()
Return output dimension of trees in a ForestContainer object
Usage
ForestSamples$leaf_dimension()
Returns
Leaf node parameter size
Method is_constant_leaf()
Return constant leaf status of trees in a ForestContainer object
Usage
ForestSamples$is_constant_leaf()
Returns
TRUE if leaves are constant, FALSE otherwise
Method is_exponentiated()
Return exponentiation status of trees in a ForestContainer object
Usage
ForestSamples$is_exponentiated()
Returns
TRUE if leaf predictions must be exponentiated, FALSE otherwise
Method add_forest_with_constant_leaves()
Add a new all-root ensemble to the container, with all of the leaves
set to the value / vector provided
Usage
ForestSamples$add_forest_with_constant_leaves(leaf_value)
Arguments
leaf_valueValue (or vector of values) to initialize root nodes in tree
Method add_numeric_split_tree()
Add a numeric (i.e. X[,i] <= c) split to a given tree in the ensemble
Usage
ForestSamples$add_numeric_split_tree(
forest_num,
tree_num,
leaf_num,
feature_num,
split_threshold,
left_leaf_value,
right_leaf_value
)
Arguments
forest_numIndex of the forest which contains the tree to be split
tree_numIndex of the tree to be split
leaf_numLeaf to be split
feature_numFeature that defines the new split
split_thresholdValue that defines the cutoff of the new split
left_leaf_valueValue (or vector of values) to assign to the newly created left node
right_leaf_valueValue (or vector of values) to assign to the newly created right node
Method get_tree_leaves()
Retrieve a vector of indices of leaf nodes for a given tree in a given forest
Usage
ForestSamples$get_tree_leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest which contains tree tree_num
tree_numIndex of the tree for which leaf indices will be retrieved
Method get_tree_split_counts()
Retrieve a vector of split counts for every training set variable in a given tree in a given forest
Usage
ForestSamples$get_tree_split_counts(forest_num, tree_num, num_features)
Arguments
forest_numIndex of the forest which contains tree tree_num
tree_numIndex of the tree for which split counts will be retrieved
num_featuresTotal number of features in the training set
Method get_forest_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest
Usage
ForestSamples$get_forest_split_counts(forest_num, num_features)
Arguments
forest_numIndex of the forest for which split counts will be retrieved
num_featuresTotal number of features in the training set
Method get_aggregate_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest, aggregated across ensembles and trees
Usage
ForestSamples$get_aggregate_split_counts(num_features)
Arguments
num_featuresTotal number of features in the training set
Method get_granular_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest, reported separately for each ensemble and tree
Usage
ForestSamples$get_granular_split_counts(num_features)
Arguments
num_featuresTotal number of features in the training set
Method ensemble_tree_max_depth()
Maximum depth of a specific tree in a specific ensemble in a ForestSamples object
Usage
ForestSamples$ensemble_tree_max_depth(ensemble_num, tree_num)
Arguments
ensemble_numEnsemble number
tree_numTree index within ensemble ensemble_num
Returns
Maximum leaf depth
Method average_ensemble_max_depth()
Average the maximum depth of each tree in a given ensemble in a ForestSamples object
Usage
ForestSamples$average_ensemble_max_depth(ensemble_num)
Arguments
ensemble_numEnsemble number
Returns
Average maximum depth
Method average_max_depth()
Average the maximum depth of each tree in each ensemble in a ForestContainer object
Usage
ForestSamples$average_max_depth()
Returns
Average maximum depth
Method num_forest_leaves()
Number of leaves in a given ensemble in a ForestSamples object
Usage
ForestSamples$num_forest_leaves(forest_num)
Arguments
forest_numIndex of the ensemble to be queried
Returns
Count of leaves in the ensemble stored at forest_num
Method sum_leaves_squared()
Sum of squared (raw) leaf values in a given ensemble in a ForestSamples object
Usage
ForestSamples$sum_leaves_squared(forest_num)
Arguments
forest_numIndex of the ensemble to be queried
Returns
Average maximum depth
Method is_leaf_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a leaf
Usage
ForestSamples$is_leaf_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a leaf, FALSE otherwise
Method is_numeric_split_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a numeric split node
Usage
ForestSamples$is_numeric_split_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a numeric split node, FALSE otherwise
Method is_categorical_split_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a categorical split node
Usage
ForestSamples$is_categorical_split_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a categorical split node, FALSE otherwise
Method parent_node()
Parent node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$parent_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the parent node
Method left_child_node()
Left child node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$left_child_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the left child node
Method right_child_node()
Right child node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$right_child_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the right child node
Method node_depth()
Depth of given node of a given tree in a given forest in a ForestSamples object, with 0 depth for the root node.
Usage
ForestSamples$node_depth(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer valued depth of the node
Method node_split_index()
Split index of given node of a given tree in a given forest in a ForestSamples object. Returns -1 is node is a leaf.
Usage
ForestSamples$node_split_index(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer valued depth of the node
Method node_split_threshold()
Threshold that defines a numeric split for a given node of a given tree in a given forest in a ForestSamples object.
Returns Inf if the node is a leaf or a categorical split node.
Usage
ForestSamples$node_split_threshold(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Threshold defining a split for the node
Method node_split_categories()
Array of category indices that define a categorical split for a given node of a given tree in a given forest in a ForestSamples object.
Returns c(Inf) if the node is a leaf or a numeric split node.
Usage
ForestSamples$node_split_categories(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Categories defining a split for the node
Method node_leaf_values()
Leaf node value(s) for a given node of a given tree in a given forest in a ForestSamples object.
Values are stale if the node is a split node.
Usage
ForestSamples$node_leaf_values(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Vector (often univariate) of leaf values
Method num_nodes()
Number of nodes in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree nodes
Method num_leaves()
Number of leaves in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree leaves
Method num_leaf_parents()
Number of leaf parents (split nodes with two leaves as children) in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_leaf_parents(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree leaf parents
Method num_split_nodes()
Number of split nodes in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_split_nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree split nodes
Method nodes()
Array of node indices in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Indices of tree nodes
Method leaves()
Array of leaf indices in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Indices of leaf nodes
Method delete_sample()
Modify the ForestSamples object by removing the forest sample indexed by 'forest_num
Usage
ForestSamples$delete_sample(forest_num)
Arguments
forest_numIndex of the forest to be removed
stochtree documentation built on April 4, 2025, 2:11 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| ForestSamples | R Documentation |
Class that stores draws from an random ensemble of decision trees
Description
Wrapper around a C++ container of tree ensembles
Public fields
forest_container_ptrExternal pointer to a C++ ForestContainer class
Methods
Public methods
Method new()
Create a new ForestContainer object.
Usage
ForestSamples$new( num_trees, leaf_dimension = 1, is_leaf_constant = FALSE, is_exponentiated = FALSE )
Arguments
num_treesNumber of trees
leaf_dimensionDimensionality of the outcome model
is_leaf_constantWhether leaf is constant
is_exponentiatedWhether forest predictions should be exponentiated before being returned
Returns
A new ForestContainer object.
Method load_from_json()
Create a new ForestContainer object from a json object
Usage
ForestSamples$load_from_json(json_object, json_forest_label)
Arguments
json_objectObject of class
CppJsonjson_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
A new ForestContainer object.
Method append_from_json()
Append to a ForestContainer object from a json object
Usage
ForestSamples$append_from_json(json_object, json_forest_label)
Arguments
json_objectObject of class
CppJsonjson_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
None
Method load_from_json_string()
Create a new ForestContainer object from a json object
Usage
ForestSamples$load_from_json_string(json_string, json_forest_label)
Arguments
json_stringJSON string which parses into object of class
CppJsonjson_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
A new ForestContainer object.
Method append_from_json_string()
Append to a ForestContainer object from a json object
Usage
ForestSamples$append_from_json_string(json_string, json_forest_label)
Arguments
json_stringJSON string which parses into object of class
CppJsonjson_forest_labelLabel referring to a particular forest (i.e. "forest_0") in the overall json hierarchy
Returns
None
Method predict()
Predict every tree ensemble on every sample in forest_dataset
Usage
ForestSamples$predict(forest_dataset)
Arguments
forest_datasetForestDatasetR class
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as samples in the ForestContainer
Method predict_raw()
Predict "raw" leaf values (without being multiplied by basis) for every tree ensemble on every sample in forest_dataset
Usage
ForestSamples$predict_raw(forest_dataset)
Arguments
forest_datasetForestDatasetR class
Returns
Array of predictions for each observation in forest_dataset and
each sample in the ForestSamples class with each prediction having the
dimensionality of the forests' leaf model. In the case of a constant leaf model
or univariate leaf regression, this array is two-dimensional (number of observations,
number of forest samples). In the case of a multivariate leaf regression,
this array is three-dimension (number of observations, leaf model dimension,
number of samples).
Method predict_raw_single_forest()
Predict "raw" leaf values (without being multiplied by basis) for a specific forest on every sample in forest_dataset
Usage
ForestSamples$predict_raw_single_forest(forest_dataset, forest_num)
Arguments
forest_datasetForestDatasetR classforest_numIndex of the forest sample within the container
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as dimensions in the leaves of trees in ForestContainer
Method predict_raw_single_tree()
Predict "raw" leaf values (without being multiplied by basis) for a specific tree in a specific forest on every observation in forest_dataset
Usage
ForestSamples$predict_raw_single_tree(forest_dataset, forest_num, tree_num)
Arguments
forest_datasetForestDatasetR classforest_numIndex of the forest sample within the container
tree_numIndex of the tree to be queried
Returns
matrix of predictions with as many rows as in forest_dataset
and as many columns as dimensions in the leaves of trees in ForestContainer
Method set_root_leaves()
Set a constant predicted value for every tree in the ensemble. Stops program if any tree is more than a root node.
Usage
ForestSamples$set_root_leaves(forest_num, leaf_value)
Arguments
forest_numIndex of the forest sample within the container.
leaf_valueConstant leaf value(s) to be fixed for each tree in the ensemble indexed by
forest_num. Can be either a single number or a vector, depending on the forest's leaf dimension.
Method prepare_for_sampler()
Set a constant predicted value for every tree in the ensemble. Stops program if any tree is more than a root node.
Usage
ForestSamples$prepare_for_sampler( dataset, outcome, forest_model, leaf_model_int, leaf_value )
Arguments
datasetForestDatasetDataset class (covariates, basis, etc...)outcomeOutcomeOutcome class (residual / partial residual)forest_modelForestModelobject storing tracking structures used in training / samplingleaf_model_intInteger value encoding the leaf model type (0 = constant gaussian, 1 = univariate gaussian, 2 = multivariate gaussian, 3 = log linear variance).
leaf_valueConstant leaf value(s) to be fixed for each tree in the ensemble indexed by
forest_num. Can be either a single number or a vector, depending on the forest's leaf dimension.
Method adjust_residual()
Adjusts residual based on the predictions of a forest
This is typically run just once at the beginning of a forest sampling algorithm. After trees are initialized with constant root node predictions, their root predictions are subtracted out of the residual.
Usage
ForestSamples$adjust_residual( dataset, outcome, forest_model, requires_basis, forest_num, add )
Arguments
datasetForestDatasetobject storing the covariates and bases for a given forestoutcomeOutcomeobject storing the residuals to be updated based on forest predictionsforest_modelForestModelobject storing tracking structures used in training / samplingrequires_basisWhether or not a forest requires a basis for prediction
forest_numIndex of forest used to update residuals
addWhether forest predictions should be added to or subtracted from residuals
Method save_json()
Store the trees and metadata of ForestDataset class in a json file
Usage
ForestSamples$save_json(json_filename)
Arguments
json_filenameName of output json file (must end in ".json")
Method load_json()
Load trees and metadata for an ensemble from a json file. Note that
any trees and metadata already present in ForestDataset class will
be overwritten.
Usage
ForestSamples$load_json(json_filename)
Arguments
json_filenameName of model input json file (must end in ".json")
Method num_samples()
Return number of samples in a ForestContainer object
Usage
ForestSamples$num_samples()
Returns
Sample count
Method num_trees()
Return number of trees in each ensemble of a ForestContainer object
Usage
ForestSamples$num_trees()
Returns
Tree count
Method leaf_dimension()
Return output dimension of trees in a ForestContainer object
Usage
ForestSamples$leaf_dimension()
Returns
Leaf node parameter size
Method is_constant_leaf()
Return constant leaf status of trees in a ForestContainer object
Usage
ForestSamples$is_constant_leaf()
Returns
TRUE if leaves are constant, FALSE otherwise
Method is_exponentiated()
Return exponentiation status of trees in a ForestContainer object
Usage
ForestSamples$is_exponentiated()
Returns
TRUE if leaf predictions must be exponentiated, FALSE otherwise
Method add_forest_with_constant_leaves()
Add a new all-root ensemble to the container, with all of the leaves set to the value / vector provided
Usage
ForestSamples$add_forest_with_constant_leaves(leaf_value)
Arguments
leaf_valueValue (or vector of values) to initialize root nodes in tree
Method add_numeric_split_tree()
Add a numeric (i.e. X[,i] <= c) split to a given tree in the ensemble
Usage
ForestSamples$add_numeric_split_tree( forest_num, tree_num, leaf_num, feature_num, split_threshold, left_leaf_value, right_leaf_value )
Arguments
forest_numIndex of the forest which contains the tree to be split
tree_numIndex of the tree to be split
leaf_numLeaf to be split
feature_numFeature that defines the new split
split_thresholdValue that defines the cutoff of the new split
left_leaf_valueValue (or vector of values) to assign to the newly created left node
right_leaf_valueValue (or vector of values) to assign to the newly created right node
Method get_tree_leaves()
Retrieve a vector of indices of leaf nodes for a given tree in a given forest
Usage
ForestSamples$get_tree_leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest which contains tree
tree_numtree_numIndex of the tree for which leaf indices will be retrieved
Method get_tree_split_counts()
Retrieve a vector of split counts for every training set variable in a given tree in a given forest
Usage
ForestSamples$get_tree_split_counts(forest_num, tree_num, num_features)
Arguments
forest_numIndex of the forest which contains tree
tree_numtree_numIndex of the tree for which split counts will be retrieved
num_featuresTotal number of features in the training set
Method get_forest_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest
Usage
ForestSamples$get_forest_split_counts(forest_num, num_features)
Arguments
forest_numIndex of the forest for which split counts will be retrieved
num_featuresTotal number of features in the training set
Method get_aggregate_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest, aggregated across ensembles and trees
Usage
ForestSamples$get_aggregate_split_counts(num_features)
Arguments
num_featuresTotal number of features in the training set
Method get_granular_split_counts()
Retrieve a vector of split counts for every training set variable in a given forest, reported separately for each ensemble and tree
Usage
ForestSamples$get_granular_split_counts(num_features)
Arguments
num_featuresTotal number of features in the training set
Method ensemble_tree_max_depth()
Maximum depth of a specific tree in a specific ensemble in a ForestSamples object
Usage
ForestSamples$ensemble_tree_max_depth(ensemble_num, tree_num)
Arguments
ensemble_numEnsemble number
tree_numTree index within ensemble
ensemble_num
Returns
Maximum leaf depth
Method average_ensemble_max_depth()
Average the maximum depth of each tree in a given ensemble in a ForestSamples object
Usage
ForestSamples$average_ensemble_max_depth(ensemble_num)
Arguments
ensemble_numEnsemble number
Returns
Average maximum depth
Method average_max_depth()
Average the maximum depth of each tree in each ensemble in a ForestContainer object
Usage
ForestSamples$average_max_depth()
Returns
Average maximum depth
Method num_forest_leaves()
Number of leaves in a given ensemble in a ForestSamples object
Usage
ForestSamples$num_forest_leaves(forest_num)
Arguments
forest_numIndex of the ensemble to be queried
Returns
Count of leaves in the ensemble stored at forest_num
Method sum_leaves_squared()
Sum of squared (raw) leaf values in a given ensemble in a ForestSamples object
Usage
ForestSamples$sum_leaves_squared(forest_num)
Arguments
forest_numIndex of the ensemble to be queried
Returns
Average maximum depth
Method is_leaf_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a leaf
Usage
ForestSamples$is_leaf_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a leaf, FALSE otherwise
Method is_numeric_split_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a numeric split node
Usage
ForestSamples$is_numeric_split_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a numeric split node, FALSE otherwise
Method is_categorical_split_node()
Whether or not a given node of a given tree in a given forest in the ForestSamples is a categorical split node
Usage
ForestSamples$is_categorical_split_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
TRUE if node is a categorical split node, FALSE otherwise
Method parent_node()
Parent node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$parent_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the parent node
Method left_child_node()
Left child node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$left_child_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the left child node
Method right_child_node()
Right child node of given node of a given tree in a given forest in a ForestSamples object
Usage
ForestSamples$right_child_node(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer ID of the right child node
Method node_depth()
Depth of given node of a given tree in a given forest in a ForestSamples object, with 0 depth for the root node.
Usage
ForestSamples$node_depth(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer valued depth of the node
Method node_split_index()
Split index of given node of a given tree in a given forest in a ForestSamples object. Returns -1 is node is a leaf.
Usage
ForestSamples$node_split_index(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Integer valued depth of the node
Method node_split_threshold()
Threshold that defines a numeric split for a given node of a given tree in a given forest in a ForestSamples object.
Returns Inf if the node is a leaf or a categorical split node.
Usage
ForestSamples$node_split_threshold(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Threshold defining a split for the node
Method node_split_categories()
Array of category indices that define a categorical split for a given node of a given tree in a given forest in a ForestSamples object.
Returns c(Inf) if the node is a leaf or a numeric split node.
Usage
ForestSamples$node_split_categories(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Categories defining a split for the node
Method node_leaf_values()
Leaf node value(s) for a given node of a given tree in a given forest in a ForestSamples object.
Values are stale if the node is a split node.
Usage
ForestSamples$node_leaf_values(forest_num, tree_num, node_id)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
node_idIndex of the node to be queried
Returns
Vector (often univariate) of leaf values
Method num_nodes()
Number of nodes in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree nodes
Method num_leaves()
Number of leaves in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree leaves
Method num_leaf_parents()
Number of leaf parents (split nodes with two leaves as children) in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_leaf_parents(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree leaf parents
Method num_split_nodes()
Number of split nodes in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$num_split_nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Count of total tree split nodes
Method nodes()
Array of node indices in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$nodes(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Indices of tree nodes
Method leaves()
Array of leaf indices in a given tree in a given forest in a ForestSamples object.
Usage
ForestSamples$leaves(forest_num, tree_num)
Arguments
forest_numIndex of the forest to be queried
tree_numIndex of the tree to be queried
Returns
Indices of leaf nodes
Method delete_sample()
Modify the ForestSamples object by removing the forest sample indexed by 'forest_num
Usage
ForestSamples$delete_sample(forest_num)
Arguments
forest_numIndex of the forest to be removed
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.