Title: Learning with Subset Stacking
Version: 0.1.0
Description: "Learning with Subset Stacking" is a supervised learning algorithm that is based on training many local estimators on subsets of a given dataset, and then passing their predictions to a global estimator. You can find the details about LESS in our manuscript at <doi:10.48550/arXiv.2112.06251>.
License: MIT + file LICENSE
Encoding: UTF-8
RoxygenNote: 7.2.1
Imports: caret, e1071, FNN, MLmetrics, pracma, R6, randomForest, RANN, rpart, wordspace
Depends: R (≥ 2.10)
LazyData: true
NeedsCompilation: no
Packaged: 2022-09-27 10:19:15 UTC; ozerc
Author: Ilker Birbil [aut], Burhan Ozer Cavdar [aut, trl, cre]
Maintainer: Burhan Ozer Cavdar <bcavdar17@ku.edu.tr>
Repository: CRAN
Date/Publication: 2022-09-27 11:00:02 UTC

BaseEstimator

Description

A dummy base R6 class that provides get_all_fields, get_attributes and set_random_state functionalities for estimators

Value

R6 Class of BaseEstimator

Methods

Public methods

Method get_all_fields()

Auxiliary function returning the name of all private and public fields of the self class

Usage
BaseEstimator$get_all_fields()
Examples
TestClass <- R6::R6Class(classname = "TestClass",
inherit = BaseEstimator,
private = list(random_state = NULL))
exampleClass <- TestClass$new()
exampleClass$get_all_fields()

Method get_attributes()

Auxiliary function returning the name and values of all private and public fields of the self class

Usage
BaseEstimator$get_attributes()
Examples
exampleClass$get_attributes()

Method set_random_state()

Auxiliary function that sets random state attribute of the self class

Usage
BaseEstimator$set_random_state(random_state)
Arguments
random_state

seed number to be set as random state

Returns

self

Examples
exampleClass$set_random_state(2022)

Method clone()

The objects of this class are cloneable with this method.

Usage
BaseEstimator$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

Examples


## ------------------------------------------------
## Method `BaseEstimator$get_all_fields`
## ------------------------------------------------

TestClass <- R6::R6Class(classname = "TestClass",
inherit = BaseEstimator,
private = list(random_state = NULL))
exampleClass <- TestClass$new()
exampleClass$get_all_fields()

## ------------------------------------------------
## Method `BaseEstimator$get_attributes`
## ------------------------------------------------

exampleClass$get_attributes()

## ------------------------------------------------
## Method `BaseEstimator$set_random_state`
## ------------------------------------------------

exampleClass$set_random_state(2022)

CoverTree - Nearest Neighbor Search

Description

Wrapper R6 Class of FNN::get.knnx function that can be used for LESSRegressor and LESSClassifier

Details

The cover tree is O(n) space data structure which allows us to answer queries in the same O(log(n)) time as kd tree given a fixed intrinsic dimensionality. Templated code from https://hunch.net/~jl/projects/cover_tree/cover_tree.html is used.

Value

R6 Class of CoverTree

Methods

Public methods

Method new()

Creates a new instance of R6 Class of CoverTree

Usage
CoverTree$new(X = NULL)
Arguments
X

An M x d data.frame or matrix, where each of the M rows is a point or a (column) vector (where d=1).

Examples
data(abalone)
ct <- CoverTree$new(abalone[1:100,])

Method query()

Finds the p number of near neighbours for each point in an input/output dataset.

Usage
CoverTree$query(query_X = private$X, k = 1)
Arguments
query_X

A set of N x d points that will be queried against X. d, the number of columns, must be the same as X. If missing, defaults to X.

k

The maximum number of nearest neighbours to compute (deafults to 1).

Returns

A list of length 2 with elements:

nn.idx A N x k integer matrix returning the near neighbour indices.
nn.dists A N x k matrix returning the near neighbour Euclidean distances
Examples
res <- ct$query(abalone[1:3,], k=2)
print(res)

Method clone()

The objects of this class are cloneable with this method.

Usage
CoverTree$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

FNN::get.knnx()

Examples


## ------------------------------------------------
## Method `CoverTree$new`
## ------------------------------------------------

data(abalone)
ct <- CoverTree$new(abalone[1:100,])

## ------------------------------------------------
## Method `CoverTree$query`
## ------------------------------------------------

res <- ct$query(abalone[1:3,], k=2)
print(res)

DecisionTreeClassifier

Description

Wrapper R6 Class of rpart::rpart function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of DecisionTreeClassifier

Super classes

less::BaseEstimator -> less::SklearnEstimator -> DecisionTreeClassifier

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of DecisionTreeClassifier

Usage
DecisionTreeClassifier$new(
  min_samples_split = 2,
  min_samples_leaf = 1,
  cp = 0.001,
  xval = 10,
  surrogate_style = 0,
  max_depth = 30
)
Arguments
min_samples_split

The minimum number of observations that must exist in a node in order for a split to be attempted (defaults to 2).

min_samples_leaf

The minimum number of observations in any terminal (leaf) node (defaults to 1).

cp

Complexity Parameter. Any split that does not decrease the overall lack of fit by a factor of cp is not attempted. This means that the overall R-squared must increase by cp at each step. The main role of this parameter is to save computing time by pruning off splits that are obviously not worthwhile. (defaults to 0.001)

xval

Number of cross-validations (defaults to 10)

surrogate_style

Controls the selection of a best surrogate. If set to 0 (default) the program uses the total number of correct classification for a potential surrogate variable, if set to 1 it uses the percent correct, calculated over the non-missing values of the surrogate. The first option more severely penalizes covariates with a large number of missing values.

max_depth

The maximum depth of any node of the final tree, with the root node counted as depth 0. Values greater than 30 will give nonsense results on 32-bit machines.

Examples
dt <- DecisionTreeClassifier$new()
dt <- DecisionTreeClassifier$new(min_samples_split = 10)
dt <- DecisionTreeClassifier$new(min_samples_leaf = 6, cp = 0.01)

Method fit()

Builds a decision tree regressor from the training set (X, y).

Usage
DecisionTreeClassifier$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes labels

Returns

Fitted R6 Class of DecisionTreeClassifier

Examples
data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

dt <- DecisionTreeClassifier$new()
dt$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
DecisionTreeClassifier$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Factor of the predict classes.

Examples
dt <- DecisionTreeClassifier$new()
dt$fit(X_train, y_train)
preds <- dt$predict(X_test)

dt <- DecisionTreeClassifier$new()
preds <- dt$fit(X_train, y_train)$predict(X_test)

preds <- DecisionTreeClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
DecisionTreeClassifier$get_estimator_type()
Examples
dt$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
DecisionTreeClassifier$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

rpart::rpart()

Examples


## ------------------------------------------------
## Method `DecisionTreeClassifier$new`
## ------------------------------------------------

dt <- DecisionTreeClassifier$new()
dt <- DecisionTreeClassifier$new(min_samples_split = 10)
dt <- DecisionTreeClassifier$new(min_samples_leaf = 6, cp = 0.01)

## ------------------------------------------------
## Method `DecisionTreeClassifier$fit`
## ------------------------------------------------

data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

dt <- DecisionTreeClassifier$new()
dt$fit(X_train, y_train)

## ------------------------------------------------
## Method `DecisionTreeClassifier$predict`
## ------------------------------------------------

dt <- DecisionTreeClassifier$new()
dt$fit(X_train, y_train)
preds <- dt$predict(X_test)

dt <- DecisionTreeClassifier$new()
preds <- dt$fit(X_train, y_train)$predict(X_test)

preds <- DecisionTreeClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

## ------------------------------------------------
## Method `DecisionTreeClassifier$get_estimator_type`
## ------------------------------------------------

dt$get_estimator_type()

DecisionTreeRegressor

Description

Wrapper R6 Class of rpart::rpart function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of DecisionTreeRegressor

Super classes

less::BaseEstimator -> less::SklearnEstimator -> DecisionTreeRegressor

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of DecisionTreeRegressor

Usage
DecisionTreeRegressor$new(
  min_samples_split = 2,
  min_samples_leaf = 1,
  cp = 0.001,
  xval = 10,
  surrogate_style = 0,
  max_depth = 30
)
Arguments
min_samples_split

The minimum number of observations that must exist in a node in order for a split to be attempted (defaults to 2).

min_samples_leaf

The minimum number of observations in any terminal (leaf) node (defaults to 1).

cp

Complexity Parameter. Any split that does not decrease the overall lack of fit by a factor of cp is not attempted. This means that the overall R-squared must increase by cp at each step. The main role of this parameter is to save computing time by pruning off splits that are obviously not worthwhile. (defaults to 0.001)

xval

Number of cross-validations (defaults to 10)

surrogate_style

Controls the selection of a best surrogate. If set to 0 (default) the program uses the total number of correct classification for a potential surrogate variable, if set to 1 it uses the percent correct, calculated over the non-missing values of the surrogate. The first option more severely penalizes covariates with a large number of missing values.

max_depth

The maximum depth of any node of the final tree, with the root node counted as depth 0. Values greater than 30 will give nonsense results on 32-bit machines.

Examples
dt <- DecisionTreeRegressor$new()
dt <- DecisionTreeRegressor$new(min_samples_split = 10)
dt <- DecisionTreeRegressor$new(min_samples_leaf = 6, cp = 0.01)

Method fit()

Builds a decision tree regressor from the training set (X, y).

Usage
DecisionTreeRegressor$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of DecisionTreeRegressor

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

dt <- DecisionTreeRegressor$new()
dt$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
DecisionTreeRegressor$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

The predict values.

Examples
dt <- DecisionTreeRegressor$new()
dt$fit(X_train, y_train)
preds <- dt$predict(X_test)

dt <- DecisionTreeRegressor$new()
preds <- dt$fit(X_train, y_train)$predict(X_test)

preds <- DecisionTreeRegressor$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
DecisionTreeRegressor$get_estimator_type()
Examples
dt$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
DecisionTreeRegressor$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

rpart::rpart()

Examples


## ------------------------------------------------
## Method `DecisionTreeRegressor$new`
## ------------------------------------------------

dt <- DecisionTreeRegressor$new()
dt <- DecisionTreeRegressor$new(min_samples_split = 10)
dt <- DecisionTreeRegressor$new(min_samples_leaf = 6, cp = 0.01)

## ------------------------------------------------
## Method `DecisionTreeRegressor$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

dt <- DecisionTreeRegressor$new()
dt$fit(X_train, y_train)

## ------------------------------------------------
## Method `DecisionTreeRegressor$predict`
## ------------------------------------------------

dt <- DecisionTreeRegressor$new()
dt$fit(X_train, y_train)
preds <- dt$predict(X_test)

dt <- DecisionTreeRegressor$new()
preds <- dt$fit(X_train, y_train)$predict(X_test)

preds <- DecisionTreeRegressor$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `DecisionTreeRegressor$get_estimator_type`
## ------------------------------------------------

dt$get_estimator_type()

Hierarchical Clustering

Description

Wrapper R6 Class of stats::hclust function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of HierarchicalClustering

Super class

less::BaseEstimator -> HierarchicalClustering

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of HierarchicalClustering

Usage
HierarchicalClustering$new(linkage = "ward.D2", n_clusters = 8)
Arguments
linkage

the agglomeration method to be used. This should be (an unambiguous abbreviation of) one of "ward.D", "ward.D2", "single", "complete", "average" (= UPGMA), "mcquitty" (= WPGMA), "median" (= WPGMC) or "centroid" (= UPGMC) (defaults to ward.D2).

n_clusters

the number of clusters (defaults to 8).

Examples
hc <- HierarchicalClustering$new()
hc <- HierarchicalClustering$new(n_clusters = 10)
hc <- HierarchicalClustering$new(n_clusters = 10, linkage = "complete")

Method fit()

Perform hierarchical clustering on a data matrix.

Usage
HierarchicalClustering$fit(X)
Arguments
X

numeric matrix of data, or an object that can be coerced to such a matrix (such as a numeric vector or a data frame with all numeric columns).

Returns

Fitted R6 class of HierarchicalClustering() that has 'labels' attribute

Examples
data(abalone)
hc <- HierarchicalClustering$new()
hc$fit(abalone[1:100,])

Method get_cluster_centers()

Auxiliary function returning the cluster centers

Usage
HierarchicalClustering$get_cluster_centers()
Examples
print(hc$get_cluster_centers())

Method get_labels()

Auxiliary function returning a vector of integers (from 1:k) indicating the cluster to which each point is allocated.

Usage
HierarchicalClustering$get_labels()
Examples
print(hc$get_labels())

Method clone()

The objects of this class are cloneable with this method.

Usage
HierarchicalClustering$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

stats::hclust()

Examples


## ------------------------------------------------
## Method `HierarchicalClustering$new`
## ------------------------------------------------

hc <- HierarchicalClustering$new()
hc <- HierarchicalClustering$new(n_clusters = 10)
hc <- HierarchicalClustering$new(n_clusters = 10, linkage = "complete")

## ------------------------------------------------
## Method `HierarchicalClustering$fit`
## ------------------------------------------------

data(abalone)
hc <- HierarchicalClustering$new()
hc$fit(abalone[1:100,])

## ------------------------------------------------
## Method `HierarchicalClustering$get_cluster_centers`
## ------------------------------------------------

print(hc$get_cluster_centers())

## ------------------------------------------------
## Method `HierarchicalClustering$get_labels`
## ------------------------------------------------

print(hc$get_labels())

KDTree - Nearest Neighbor Search

Description

Wrapper R6 Class of RANN::nn2 function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of KDTree

Methods

Public methods

Method new()

Creates a new instance of R6 Class of KDTree

Usage
KDTree$new(X = NULL)
Arguments
X

An M x d data.frame or matrix, where each of the M rows is a point or a (column) vector (where d=1).

Examples
data(abalone)
kdt <- KDTree$new(abalone[1:100,])

Method query()

Finds the p number of near neighbours for each point in an input/output dataset. The advantage of the kd-tree is that it runs in O(M log M) time.

Usage
KDTree$query(query_X = private$X, k = 1)
Arguments
query_X

A set of N x d points that will be queried against X. d, the number of columns, must be the same as X. If missing, defaults to X.

k

The maximum number of nearest neighbours to compute (deafults to 1).

Returns

A list of length 2 with elements:

nn.idx A N x k integer matrix returning the near neighbour indices.
nn.dists A N x k matrix returning the near neighbour Euclidean distances
Examples
res <- kdt$query(abalone[1:3,], k=2)
print(res)

Method clone()

The objects of this class are cloneable with this method.

Usage
KDTree$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

RANN::nn2()

Examples


## ------------------------------------------------
## Method `KDTree$new`
## ------------------------------------------------

data(abalone)
kdt <- KDTree$new(abalone[1:100,])

## ------------------------------------------------
## Method `KDTree$query`
## ------------------------------------------------

res <- kdt$query(abalone[1:3,], k=2)
print(res)

KMeans Clustering

Description

Wrapper R6 Class of stats::kmeans function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of KMeans

Super class

less::BaseEstimator -> KMeans

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of KMeans

Usage
KMeans$new(n_clusters = 8, n_init = 10, max_iter = 300, random_state = NULL)
Arguments
n_clusters

the number of clusters. A random set of (distinct) rows in X is chosen as the initial centres (default to 8)

n_init

how many random sets should be chosen? (default to 10)

max_iter

the maximum number of iterations allowed (default to 300).

random_state

seed number to be used for fixing the randomness (default to NULL).

Examples
km <- KMeans$new()
km <- KMeans$new(n_clusters = 10)
km <- KMeans$new(n_clusters = 10, random_state = 100)

Method fit()

Perform k-means clustering on a data matrix.

Usage
KMeans$fit(X)
Arguments
X

numeric matrix of data, or an object that can be coerced to such a matrix (such as a numeric vector or a data frame with all numeric columns).

Returns

Fitted R6 class of KMeans() that has 'cluster_centers' and 'labels' attributes

Examples
data(abalone)
km <- KMeans$new()
km$fit(abalone[1:100,])

Method get_cluster_centers()

Auxiliary function returning the cluster centers

Usage
KMeans$get_cluster_centers()
Examples
print(km$get_cluster_centers())

Method get_labels()

Auxiliary function returning a vector of integers (from 1:k) indicating the cluster to which each point is allocated.

Usage
KMeans$get_labels()
Examples
print(km$get_labels())

Method clone()

The objects of this class are cloneable with this method.

Usage
KMeans$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

stats::kmeans()

Examples


## ------------------------------------------------
## Method `KMeans$new`
## ------------------------------------------------

km <- KMeans$new()
km <- KMeans$new(n_clusters = 10)
km <- KMeans$new(n_clusters = 10, random_state = 100)

## ------------------------------------------------
## Method `KMeans$fit`
## ------------------------------------------------

data(abalone)
km <- KMeans$new()
km$fit(abalone[1:100,])

## ------------------------------------------------
## Method `KMeans$get_cluster_centers`
## ------------------------------------------------

print(km$get_cluster_centers())

## ------------------------------------------------
## Method `KMeans$get_labels`
## ------------------------------------------------

print(km$get_labels())

KNeighborsClassifier

Description

Wrapper R6 Class of caret::knnreg function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of KNeighborsClassifier

Super classes

less::BaseEstimator -> less::SklearnEstimator -> KNeighborsClassifier

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of KNeighborsClassifier

Usage
KNeighborsClassifier$new(k = 5)
Arguments
k

Number of neighbors considered (defaults to 5).

Examples
knc <- KNeighborsClassifier$new()
knc <- KNeighborsClassifier$new(k = 5)

Method fit()

Fit the k-nearest neighbors regressor from the training set (X, y).

Usage
KNeighborsClassifier$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes labels

Returns

Fitted R6 Class of KNeighborsClassifier

Examples
data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

knc <- KNeighborsClassifier$new()
knc$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
KNeighborsClassifier$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Factor of the predict classes.

Examples
knc <- KNeighborsClassifier$new()
knc$fit(X_train, y_train)
preds <- knc$predict(X_test)

knc <- KNeighborsClassifier$new()
preds <- knc$fit(X_train, y_train)$predict(X_test)

preds <- KNeighborsClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=factor(preds), reference = factor(y_test)))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
KNeighborsClassifier$get_estimator_type()
Examples
knc$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
KNeighborsClassifier$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

caret::knn3()

Examples


## ------------------------------------------------
## Method `KNeighborsClassifier$new`
## ------------------------------------------------

knc <- KNeighborsClassifier$new()
knc <- KNeighborsClassifier$new(k = 5)

## ------------------------------------------------
## Method `KNeighborsClassifier$fit`
## ------------------------------------------------

data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

knc <- KNeighborsClassifier$new()
knc$fit(X_train, y_train)

## ------------------------------------------------
## Method `KNeighborsClassifier$predict`
## ------------------------------------------------

knc <- KNeighborsClassifier$new()
knc$fit(X_train, y_train)
preds <- knc$predict(X_test)

knc <- KNeighborsClassifier$new()
preds <- knc$fit(X_train, y_train)$predict(X_test)

preds <- KNeighborsClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=factor(preds), reference = factor(y_test)))

## ------------------------------------------------
## Method `KNeighborsClassifier$get_estimator_type`
## ------------------------------------------------

knc$get_estimator_type()

KNeighborsRegressor

Description

Wrapper R6 Class of caret::knnreg function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of KNeighborsRegressor

Super classes

less::BaseEstimator -> less::SklearnEstimator -> KNeighborsRegressor

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of KNeighborsRegressor

Usage
KNeighborsRegressor$new(k = 5)
Arguments
k

Number of neighbors considered (defaults to 5).

Examples
knr <- KNeighborsRegressor$new()
knr <- KNeighborsRegressor$new(k = 5)

Method fit()

Fit the k-nearest neighbors regressor from the training set (X, y).

Usage
KNeighborsRegressor$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of KNeighborsRegressor

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

knr <- KNeighborsRegressor$new()
knr$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
KNeighborsRegressor$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

The predict values.

Examples
knr <- KNeighborsRegressor$new()
knr$fit(X_train, y_train)
preds <- knr$predict(X_test)

knr <- KNeighborsRegressor$new()
preds <- knr$fit(X_train, y_train)$predict(X_test)

preds <- KNeighborsRegressor$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
KNeighborsRegressor$get_estimator_type()
Examples
knr$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
KNeighborsRegressor$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

caret::knnreg()

Examples


## ------------------------------------------------
## Method `KNeighborsRegressor$new`
## ------------------------------------------------

knr <- KNeighborsRegressor$new()
knr <- KNeighborsRegressor$new(k = 5)

## ------------------------------------------------
## Method `KNeighborsRegressor$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

knr <- KNeighborsRegressor$new()
knr$fit(X_train, y_train)

## ------------------------------------------------
## Method `KNeighborsRegressor$predict`
## ------------------------------------------------

knr <- KNeighborsRegressor$new()
knr$fit(X_train, y_train)
preds <- knr$predict(X_test)

knr <- KNeighborsRegressor$new()
preds <- knr$fit(X_train, y_train)$predict(X_test)

preds <- KNeighborsRegressor$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `KNeighborsRegressor$get_estimator_type`
## ------------------------------------------------

knr$get_estimator_type()

LESSBase

Description

The base class for LESSRegressor and LESSClassifier

Value

R6 class of LESSBase

Super classes

less::BaseEstimator -> less::SklearnEstimator -> LESSBase

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of LESSBase

Usage
LESSBase$new(replications = NULL, scobject = NULL, isFitted = FALSE)
Arguments
replications

List to store the replications

scobject

Scaling object used for normalization (less::StandardScaler)

isFitted

Flag to check whether LESS is fitted

Method set_random_state()

Auxiliary function that sets random state attribute of the self class

Usage
LESSBase$set_random_state(random_state)
Arguments
random_state

seed number to be set as random state

Returns

self

Method get_n_subsets()

Auxiliary function returning the number of subsets

Usage
LESSBase$get_n_subsets()

Method get_n_neighbors()

Auxiliary function returning the number of neighbors

Usage
LESSBase$get_n_neighbors()

Method get_frac()

Auxiliary function returning the percentage of samples used to set the number of neighbors

Usage
LESSBase$get_frac()

Method get_n_replications()

Auxiliary function returning the number of replications

Usage
LESSBase$get_n_replications()

Method get_d_normalize()

Auxiliary function returning the flag for normalization

Usage
LESSBase$get_d_normalize()

Method get_scaling()

Auxiliary function returning the flag for scaling

Usage
LESSBase$get_scaling()

Method get_val_size()

Auxiliary function returning the validation set size

Usage
LESSBase$get_val_size()

Method get_random_state()

Auxiliary function returning the random seed

Usage
LESSBase$get_random_state()

Method get_isFitted()

Auxiliary function returning the isFitted flag

Usage
LESSBase$get_isFitted()

Method get_replications()

Auxiliary function returning the isFitted flag

Usage
LESSBase$get_replications()

Method clone()

The objects of this class are cloneable with this method.

Usage
LESSBase$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

LESSBinaryClassifier

Description

Auxiliary binary classifier for Learning with Subset Stacking (LESS)

Value

R6 class of LESSBinaryClassifier

Super classes

less::BaseEstimator -> less::SklearnEstimator -> less::LESSBase -> LESSBinaryClassifier

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of LESSBinaryClassifier

Usage
LESSBinaryClassifier$new(
  frac = NULL,
  n_neighbors = NULL,
  n_subsets = NULL,
  n_replications = 20,
  d_normalize = TRUE,
  val_size = NULL,
  random_state = NULL,
  tree_method = function(X) KDTree$new(X),
  cluster_method = NULL,
  local_estimator = LinearRegression$new(),
  global_estimator = DecisionTreeClassifier$new(),
  distance_function = NULL,
  scaling = TRUE,
  warnings = TRUE
)
Arguments
frac

fraction of total samples used for the number of neighbors (default is 0.05)

n_neighbors

number of neighbors (default is NULL)

n_subsets

number of subsets (default is NULL)

n_replications

number of replications (default is 20)

d_normalize

distance normalization (default is TRUE)

val_size

percentage of samples used for validation (default is NULL - no validation)

random_state

initialization of the random seed (default is NULL)

tree_method

method used for constructing the nearest neighbor tree, e.g., less::KDTree (default)

cluster_method

method used for clustering the subsets, e.g., less::KMeans (default is NULL)

local_estimator

estimator for the local models (default is less::LinearRegression)

global_estimator

estimator for the global model (default is less::DecisionTreeRegressor)

distance_function

distance function evaluating the distance from a subset to a sample, e.g., df(subset, sample) which returns a vector of distances (default is RBF(subset, sample, 1.0/n_subsets^2))

scaling

flag to normalize the input data (default is TRUE)

warnings

flag to turn on (TRUE) or off (FALSE) the warnings (default is TRUE)

Method fit()

Dummy fit function that calls the proper method according to validation and clustering parameters Options are:

Usage
LESSBinaryClassifier$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of LESSBinaryClassifier

Method predict_proba()

Prediction probabilities are evaluated for the test samples in X0

Usage
LESSBinaryClassifier$predict_proba(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Method get_global_estimator()

Auxiliary function returning the global_estimator

Usage
LESSBinaryClassifier$get_global_estimator()

Method set_random_state()

Auxiliary function that sets random state attribute of the self class

Usage
LESSBinaryClassifier$set_random_state(random_state)
Arguments
random_state

seed number to be set as random state

Returns

self

Method clone()

The objects of this class are cloneable with this method.

Usage
LESSBinaryClassifier$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

LESSClassifier

Description

Classifier for Learning with Subset Stacking (LESS)

Value

R6 class of LESSClassifier

Super classes

less::BaseEstimator -> less::SklearnEstimator -> less::LESSBase -> LESSClassifier

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of LESSClassifier

Usage
LESSClassifier$new(
  frac = NULL,
  n_neighbors = NULL,
  n_subsets = NULL,
  n_replications = 20,
  d_normalize = TRUE,
  val_size = NULL,
  random_state = NULL,
  tree_method = function(X) KDTree$new(X),
  cluster_method = NULL,
  local_estimator = LinearRegression$new(),
  global_estimator = DecisionTreeClassifier$new(),
  distance_function = NULL,
  scaling = TRUE,
  warnings = TRUE,
  multiclass = "ovr"
)
Arguments
frac

fraction of total samples used for the number of neighbors (default is 0.05)

n_neighbors

number of neighbors (default is NULL)

n_subsets

number of subsets (default is NULL)

n_replications

number of replications (default is 20)

d_normalize

distance normalization (default is TRUE)

val_size

percentage of samples used for validation (default is NULL - no validation)

random_state

initialization of the random seed (default is NULL)

tree_method

method used for constructing the nearest neighbor tree, e.g., less::KDTree (default)

cluster_method

method used for clustering the subsets, e.g., less::KMeans (default is NULL)

local_estimator

estimator for the local models (default is less::LinearRegression)

global_estimator

estimator for the global model (default is less::DecisionTreeRegressor)

distance_function

distance function evaluating the distance from a subset to a sample, e.g., df(subset, sample) which returns a vector of distances (default is RBF(subset, sample, 1.0/n_subsets^2))

scaling

flag to normalize the input data (default is TRUE)

warnings

flag to turn on (TRUE) or off (FALSE) the warnings (default is TRUE)

multiclass

available strategies are 'ovr' (one-vs-rest, default), 'ovo' (one-vs-one), 'occ' (output-code-classifier) (default is 'ovr')

Examples
lessclassifier <- LESSClassifier$new()
lessclassifier <- LESSClassifier$new(multiclass = "ovo")

Method fit()

Dummy fit function that calls the fit method of the multiclass strategy

Usage
LESSClassifier$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of LESSClassifier

Examples
data(iris)
set.seed(2022)
shuffled_iris <- iris[sample(1:nrow(iris)),]
split_list <- train_test_split(shuffled_iris[1:10,], test_size =  0.3, random_state = 1)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lessclassifier <- LESSClassifier$new()
lessclassifier$fit(X_train, y_train)

Method predict()

Dummy predict function that calls the predict method of the multiclass strategy

Usage
LESSClassifier$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Predicted values of the given predictors

Examples
preds <- lessclassifier$predict(X_test)
print(caret::confusionMatrix(data=factor(preds), reference = factor(y_test)))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
LESSClassifier$get_estimator_type()
Examples
lessclassifier$get_estimator_type()

Method set_random_state()

Auxiliary function that sets random state attribute of the self class

Usage
LESSClassifier$set_random_state(random_state)
Arguments
random_state

seed number to be set as random state

Returns

self

Examples
lessclassifier$set_random_state(2022)

Method clone()

The objects of this class are cloneable with this method.

Usage
LESSClassifier$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

Examples


## ------------------------------------------------
## Method `LESSClassifier$new`
## ------------------------------------------------

lessclassifier <- LESSClassifier$new()
lessclassifier <- LESSClassifier$new(multiclass = "ovo")

## ------------------------------------------------
## Method `LESSClassifier$fit`
## ------------------------------------------------

data(iris)
set.seed(2022)
shuffled_iris <- iris[sample(1:nrow(iris)),]
split_list <- train_test_split(shuffled_iris[1:10,], test_size =  0.3, random_state = 1)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lessclassifier <- LESSClassifier$new()
lessclassifier$fit(X_train, y_train)

## ------------------------------------------------
## Method `LESSClassifier$predict`
## ------------------------------------------------

preds <- lessclassifier$predict(X_test)
print(caret::confusionMatrix(data=factor(preds), reference = factor(y_test)))

## ------------------------------------------------
## Method `LESSClassifier$get_estimator_type`
## ------------------------------------------------

lessclassifier$get_estimator_type()

## ------------------------------------------------
## Method `LESSClassifier$set_random_state`
## ------------------------------------------------

lessclassifier$set_random_state(2022)

LESSRegressor

Description

Regressor for Learning with Subset Stacking (LESS)

Value

R6 class of LESSRegressor

Super classes

less::BaseEstimator -> less::SklearnEstimator -> less::LESSBase -> LESSRegressor

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of LESSRegressor

Usage
LESSRegressor$new(
  frac = NULL,
  n_neighbors = NULL,
  n_subsets = NULL,
  n_replications = 20,
  d_normalize = TRUE,
  val_size = NULL,
  random_state = NULL,
  tree_method = function(X) KDTree$new(X),
  cluster_method = NULL,
  local_estimator = LinearRegression$new(),
  global_estimator = DecisionTreeRegressor$new(),
  distance_function = NULL,
  scaling = TRUE,
  warnings = TRUE
)
Arguments
frac

fraction of total samples used for the number of neighbors (default is 0.05)

n_neighbors

number of neighbors (default is NULL)

n_subsets

number of subsets (default is NULL)

n_replications

number of replications (default is 20)

d_normalize

distance normalization (default is TRUE)

val_size

percentage of samples used for validation (default is NULL - no validation)

random_state

initialization of the random seed (default is NULL)

tree_method

method used for constructing the nearest neighbor tree, e.g., less::KDTree (default)

cluster_method

method used for clustering the subsets, e.g., less::KMeans (default is NULL)

local_estimator

estimator for the local models (default is less::LinearRegression)

global_estimator

estimator for the global model (default is less::DecisionTreeRegressor)

distance_function

distance function evaluating the distance from a subset to a sample, e.g., df(subset, sample) which returns a vector of distances (default is RBF(subset, sample, 1.0/n_subsets^2))

scaling

flag to normalize the input data (default is TRUE)

warnings

flag to turn on (TRUE) or off (FALSE) the warnings (default is TRUE)

Examples
lessRegressor <- LESSRegressor$new()
lessRegressor <- LESSRegressor$new(val_size = 0.3)
lessRegressor <- LESSRegressor$new(cluster_method = less::KMeans$new())
lessRegressor <- LESSRegressor$new(val_size = 0.3, cluster_method = less::KMeans$new())

Method fit()

Dummy fit function that calls the proper method according to validation and clustering parameters Options are:

Usage
LESSRegressor$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of LESSRegressor

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lessRegressor <- LESSRegressor$new()
lessRegressor$fit(X_train, y_train)

Method predict()

Predictions are evaluated for the test samples in X0

Usage
LESSRegressor$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Predicted values of the given predictors

Examples
preds <- lessRegressor$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
LESSRegressor$get_estimator_type()
Examples
lessRegressor$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
LESSRegressor$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

LESSBase

Examples


## ------------------------------------------------
## Method `LESSRegressor$new`
## ------------------------------------------------

lessRegressor <- LESSRegressor$new()
lessRegressor <- LESSRegressor$new(val_size = 0.3)
lessRegressor <- LESSRegressor$new(cluster_method = less::KMeans$new())
lessRegressor <- LESSRegressor$new(val_size = 0.3, cluster_method = less::KMeans$new())

## ------------------------------------------------
## Method `LESSRegressor$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lessRegressor <- LESSRegressor$new()
lessRegressor$fit(X_train, y_train)

## ------------------------------------------------
## Method `LESSRegressor$predict`
## ------------------------------------------------

preds <- lessRegressor$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `LESSRegressor$get_estimator_type`
## ------------------------------------------------

lessRegressor$get_estimator_type()

LinearRegression

Description

Wrapper R6 Class of stats::lm function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of LinearRegression

Super classes

less::BaseEstimator -> less::SklearnEstimator -> LinearRegression

Methods

Public methods

Inherited methods

Method fit()

Fits a linear model (y ~ X)

Usage
LinearRegression$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of LinearRegression

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lr <- LinearRegression$new()
lr$fit(X_train, y_train)

Method predict()

Predict regression value for X.

Usage
LinearRegression$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

The predict values.

Examples
lr <- LinearRegression$new()
lr$fit(X_train, y_train)
preds <- lr$predict(X_test)

lr <- LinearRegression$new()
preds <- lr$fit(X_train, y_train)$predict(X_test)

preds <- LinearRegression$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
LinearRegression$get_estimator_type()
Examples
lr$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
LinearRegression$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

stats::lm()

Examples


## ------------------------------------------------
## Method `LinearRegression$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

lr <- LinearRegression$new()
lr$fit(X_train, y_train)

## ------------------------------------------------
## Method `LinearRegression$predict`
## ------------------------------------------------

lr <- LinearRegression$new()
lr$fit(X_train, y_train)
preds <- lr$predict(X_test)

lr <- LinearRegression$new()
preds <- lr$fit(X_train, y_train)$predict(X_test)

preds <- LinearRegression$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `LinearRegression$get_estimator_type`
## ------------------------------------------------

lr$get_estimator_type()

RandomForestClassifier

Description

Wrapper R6 Class of randomForest::randomForest function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of RandomForestClassifier

Super classes

less::BaseEstimator -> less::SklearnEstimator -> RandomForestClassifier

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of RandomForestClassifier

Usage
RandomForestClassifier$new(
  n_estimators = 100,
  random_state = NULL,
  min_samples_leaf = 1,
  max_leaf_nodes = NULL
)
Arguments
n_estimators

Number of trees to grow. This should not be set to too small a number, to ensure that every input row gets predicted at least a few times (defaults to 100).

random_state

Seed number to be used for fixing the randomness (default to NULL).

min_samples_leaf

Minimum size of terminal nodes. Setting this number larger causes smaller trees to be grown (and thus take less time) (defaults to 1)

max_leaf_nodes

Maximum number of terminal nodes trees in the forest can have. If not given, trees are grown to the maximum possible (subject to limits by nodesize). If set larger than maximum possible, a warning is issued. (defaults to NULL)

Examples
rf <- RandomForestClassifier$new()
rf <- RandomForestClassifier$new(n_estimators = 500)
rf <- RandomForestClassifier$new(n_estimators = 500, random_state = 100)

Method fit()

Builds a random forest regressor from the training set (X, y).

Usage
RandomForestClassifier$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes labels

Returns

Fitted R6 Class of RandomForestClassifier

Examples
data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

rf <- RandomForestClassifier$new()
rf$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
RandomForestClassifier$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Factor of the predict classes.

Examples
rf <- RandomForestClassifier$new()
rf$fit(X_train, y_train)
preds <- rf$predict(X_test)

rf <- RandomForestClassifier$new()
preds <- rf$fit(X_train, y_train)$predict(X_test)

preds <- RandomForestClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
RandomForestClassifier$get_estimator_type()
Examples
rf$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
RandomForestClassifier$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

randomForest::randomForest()

Examples


## ------------------------------------------------
## Method `RandomForestClassifier$new`
## ------------------------------------------------

rf <- RandomForestClassifier$new()
rf <- RandomForestClassifier$new(n_estimators = 500)
rf <- RandomForestClassifier$new(n_estimators = 500, random_state = 100)

## ------------------------------------------------
## Method `RandomForestClassifier$fit`
## ------------------------------------------------

data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

rf <- RandomForestClassifier$new()
rf$fit(X_train, y_train)

## ------------------------------------------------
## Method `RandomForestClassifier$predict`
## ------------------------------------------------

rf <- RandomForestClassifier$new()
rf$fit(X_train, y_train)
preds <- rf$predict(X_test)

rf <- RandomForestClassifier$new()
preds <- rf$fit(X_train, y_train)$predict(X_test)

preds <- RandomForestClassifier$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

## ------------------------------------------------
## Method `RandomForestClassifier$get_estimator_type`
## ------------------------------------------------

rf$get_estimator_type()

RandomForestRegressor

Description

Wrapper R6 Class of randomForest::randomForest function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of RandomForestRegressor

Super classes

less::BaseEstimator -> less::SklearnEstimator -> RandomForestRegressor

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of RandomForestRegressor

Usage
RandomForestRegressor$new(
  n_estimators = 100,
  random_state = NULL,
  min_samples_leaf = 1,
  max_leaf_nodes = NULL
)
Arguments
n_estimators

Number of trees to grow. This should not be set to too small a number, to ensure that every input row gets predicted at least a few times (defaults to 100).

random_state

Seed number to be used for fixing the randomness (default to NULL).

min_samples_leaf

Minimum size of terminal nodes. Setting this number larger causes smaller trees to be grown (and thus take less time) (defaults to 1)

max_leaf_nodes

Maximum number of terminal nodes trees in the forest can have. If not given, trees are grown to the maximum possible (subject to limits by nodesize). If set larger than maximum possible, a warning is issued. (defaults to NULL)

Examples
rf <- RandomForestRegressor$new()
rf <- RandomForestRegressor$new(n_estimators = 500)
rf <- RandomForestRegressor$new(n_estimators = 500, random_state = 100)

Method fit()

Builds a random forest regressor from the training set (X, y).

Usage
RandomForestRegressor$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of RandomForestRegressor

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

rf <- RandomForestRegressor$new()
rf$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
RandomForestRegressor$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

The predict values.

Examples
preds <- rf$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
RandomForestRegressor$get_estimator_type()
Examples
rf$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
RandomForestRegressor$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

randomForest::randomForest()

Examples


## ------------------------------------------------
## Method `RandomForestRegressor$new`
## ------------------------------------------------

rf <- RandomForestRegressor$new()
rf <- RandomForestRegressor$new(n_estimators = 500)
rf <- RandomForestRegressor$new(n_estimators = 500, random_state = 100)

## ------------------------------------------------
## Method `RandomForestRegressor$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

rf <- RandomForestRegressor$new()
rf$fit(X_train, y_train)

## ------------------------------------------------
## Method `RandomForestRegressor$predict`
## ------------------------------------------------

preds <- rf$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `RandomForestRegressor$get_estimator_type`
## ------------------------------------------------

rf$get_estimator_type()

Support Vector Classification

Description

Wrapper R6 Class of e1071::svm function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of SVC

Super classes

less::BaseEstimator -> less::SklearnEstimator -> SVC

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of SVC

Usage
SVC$new(
  scale = TRUE,
  kernel = "radial",
  degree = 3,
  gamma = NULL,
  coef0 = 0,
  cost = 1,
  cache_size = 40,
  tolerance = 0.001,
  epsilon = 0.1,
  shrinking = TRUE,
  cross = 0,
  probability = FALSE,
  fitted = TRUE
)
Arguments
scale

A logical vector indicating the variables to be scaled. If scale is of length 1, the value is recycled as many times as needed. Per default, data are scaled internally (both x and y variables) to zero mean and unit variance. The center and scale values are returned and used for later predictions (default: TRUE)

kernel

The kernel used in training and predicting. Possible values are: "linear", "polynomial", "radial", "sigmoid" (default is "radial")

degree

Parameter needed for kernel of type polynomial (default: 3)

gamma

Parameter needed for all kernels except linear (default: 1/(data dimension))

coef0

Parameter needed for kernels of type polynomial and sigmoid (default: 0)

cost

Cost of constraints violation (default: 1)—it is the ‘C’-constant of the regularization term in the Lagrange formulation (default: 1)

cache_size

Cache memory in MB (default: 40)

tolerance

Tolerance of termination criterion (default: 0.001)

epsilon

Epsilon in the insensitive-loss function (default: 0.1)

shrinking

Option whether to use the shrinking-heuristics (default: TRUE)

cross

If a integer value k>0 is specified, a k-fold cross validation on the training data is performed to assess the quality of the model: the accuracy rate for classification and the Mean Squared Error for regression (default: 0)

probability

Logical indicating whether the model should allow for probability predictions (default: FALSE)

fitted

Logical indicating whether the fitted values should be computed and included in the model or not (default: TRUE)

Examples
svc <- SVC$new()
svc <- SVC$new(kernel = "polynomial")

Method fit()

Fit the SVM model from the training set (X, y).

Usage
SVC$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes labels

Returns

Fitted R6 Class of SVC

Examples
data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

svc <- SVC$new()
svc$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
SVC$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

Factor of the predict classes.

Examples
svc <- SVC$new()
svc$fit(X_train, y_train)
preds <- svc$predict(X_test)

svc <- SVC$new()
preds <- svc$fit(X_train, y_train)$predict(X_test)

preds <- SVC$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
SVC$get_estimator_type()
Examples
svc$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
SVC$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

e1071::svm()

Examples


## ------------------------------------------------
## Method `SVC$new`
## ------------------------------------------------

svc <- SVC$new()
svc <- SVC$new(kernel = "polynomial")

## ------------------------------------------------
## Method `SVC$fit`
## ------------------------------------------------

data(iris)
split_list <- train_test_split(iris, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

svc <- SVC$new()
svc$fit(X_train, y_train)

## ------------------------------------------------
## Method `SVC$predict`
## ------------------------------------------------

svc <- SVC$new()
svc$fit(X_train, y_train)
preds <- svc$predict(X_test)

svc <- SVC$new()
preds <- svc$fit(X_train, y_train)$predict(X_test)

preds <- SVC$new()$fit(X_train, y_train)$predict(X_test)
print(caret::confusionMatrix(data=preds, reference = factor(y_test)))

## ------------------------------------------------
## Method `SVC$get_estimator_type`
## ------------------------------------------------

svc$get_estimator_type()

Support Vector Regression

Description

Wrapper R6 Class of e1071::svm function that can be used for LESSRegressor and LESSClassifier

Value

R6 Class of SVR

Super classes

less::BaseEstimator -> less::SklearnEstimator -> SVR

Methods

Public methods

Inherited methods

Method new()

Creates a new instance of R6 Class of SVR

Usage
SVR$new(
  scale = TRUE,
  kernel = "radial",
  degree = 3,
  gamma = NULL,
  coef0 = 0,
  cost = 1,
  cache_size = 40,
  tolerance = 0.001,
  epsilon = 0.1,
  shrinking = TRUE,
  cross = 0,
  probability = FALSE,
  fitted = TRUE
)
Arguments
scale

A logical vector indicating the variables to be scaled. If scale is of length 1, the value is recycled as many times as needed. Per default, data are scaled internally (both x and y variables) to zero mean and unit variance. The center and scale values are returned and used for later predictions (default: TRUE)

kernel

The kernel used in training and predicting. Possible values are: "linear", "polynomial", "radial", "sigmoid" (default is "radial")

degree

Parameter needed for kernel of type polynomial (default: 3)

gamma

Parameter needed for all kernels except linear (default: 1/(data dimension))

coef0

Parameter needed for kernels of type polynomial and sigmoid (default: 0)

cost

Cost of constraints violation (default: 1)—it is the ‘C’-constant of the regularization term in the Lagrange formulation (default: 1)

cache_size

Cache memory in MB (default: 40)

tolerance

Tolerance of termination criterion (default: 0.001)

epsilon

Epsilon in the insensitive-loss function (default: 0.1)

shrinking

Option whether to use the shrinking-heuristics (default: TRUE)

cross

If a integer value k>0 is specified, a k-fold cross validation on the training data is performed to assess the quality of the model: the accuracy rate for classification and the Mean Squared Error for regression (default: 0)

probability

Logical indicating whether the model should allow for probability predictions (default: FALSE)

fitted

Logical indicating whether the fitted values should be computed and included in the model or not (default: TRUE)

Examples
svr <- SVR$new()
svr <- SVR$new(kernel = "polynomial")

Method fit()

Fit the SVM model from the training set (X, y).

Usage
SVR$fit(X, y)
Arguments
X

2D matrix or dataframe that includes predictors

y

1D vector or (n,1) dimensional matrix/dataframe that includes response variables

Returns

Fitted R6 Class of SVR

Examples
data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

svr <- SVR$new()
svr$fit(X_train, y_train)

Method predict()

Predict regression value for X0.

Usage
SVR$predict(X0)
Arguments
X0

2D matrix or dataframe that includes predictors

Returns

The predict values.

Examples
svr <- SVR$new()
svr$fit(X_train, y_train)
preds <- svr$predict(X_test)

svr <- SVR$new()
preds <- svr$fit(X_train, y_train)$predict(X_test)

preds <- SVR$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

Method get_estimator_type()

Auxiliary function returning the estimator type e.g 'regressor', 'classifier'

Usage
SVR$get_estimator_type()
Examples
svr$get_estimator_type()

Method clone()

The objects of this class are cloneable with this method.

Usage
SVR$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

See Also

e1071::svm()

Examples


## ------------------------------------------------
## Method `SVR$new`
## ------------------------------------------------

svr <- SVR$new()
svr <- SVR$new(kernel = "polynomial")

## ------------------------------------------------
## Method `SVR$fit`
## ------------------------------------------------

data(abalone)
split_list <- train_test_split(abalone[1:100,], test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

svr <- SVR$new()
svr$fit(X_train, y_train)

## ------------------------------------------------
## Method `SVR$predict`
## ------------------------------------------------

svr <- SVR$new()
svr$fit(X_train, y_train)
preds <- svr$predict(X_test)

svr <- SVR$new()
preds <- svr$fit(X_train, y_train)$predict(X_test)

preds <- SVR$new()$fit(X_train, y_train)$predict(X_test)
print(head(matrix(c(y_test, preds), ncol = 2, dimnames = (list(NULL, c("True", "Prediction"))))))

## ------------------------------------------------
## Method `SVR$get_estimator_type`
## ------------------------------------------------

svr$get_estimator_type()

SklearnEstimator

Description

A dummy base R6 class that includes fit, predict functions for estimators

Value

R6 Class of SklearnEstimator

Super class

less::BaseEstimator -> SklearnEstimator

Methods

Public methods

Inherited methods

Method fit()

Dummy fit function

Usage
SklearnEstimator$fit()
Examples
sklearn <- SklearnEstimator$new()
sklearn$fit()

Method predict()

Dummy predict function

Usage
SklearnEstimator$predict()
Examples
sklearn$predict()

Method get_type()

Auxiliary function returning the type of the class e.g 'estimator'

Usage
SklearnEstimator$get_type()
Examples
sklearn$get_type()

Method get_isFitted()

Auxiliary function returning the isFitted flag

Usage
SklearnEstimator$get_isFitted()
Examples
sklearn$get_isFitted()

Method clone()

The objects of this class are cloneable with this method.

Usage
SklearnEstimator$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

Examples


## ------------------------------------------------
## Method `SklearnEstimator$fit`
## ------------------------------------------------

sklearn <- SklearnEstimator$new()
sklearn$fit()

## ------------------------------------------------
## Method `SklearnEstimator$predict`
## ------------------------------------------------

sklearn$predict()

## ------------------------------------------------
## Method `SklearnEstimator$get_type`
## ------------------------------------------------

sklearn$get_type()

## ------------------------------------------------
## Method `SklearnEstimator$get_isFitted`
## ------------------------------------------------

sklearn$get_isFitted()

Abalone Data Set

Description

The number of rings is the value to predict.

Usage

abalone

Format

A dataframe with 4177 rows and 8 variables

Length

Longest shell measurement in mm

Diameter

perpendicular to length in mm

Height

with meat in shell in mm

Whole weight

whole abalone in grams

Shucked weight

weight of meat in grams

Viscera weight

gut weight (after bleeding) in grams

Shell weight

after being dried in grams

Rings

+1.5 gives the age in years

Source

Dua, D. and Graff, C. (2019). UCI Machine Learning Repository http://archive.ics.uci.edu/ml/. Irvine, CA: University of California, School of Information and Computer Science.

Examples

data(abalone)

Fitting Control

Description

Checks if the given estimator is fitted

Usage

check_is_fitted(estimator)

Arguments

estimator

An estimator with is_fitted attribute

Value

TRUE if the estimator is fitted, FALSE is not

Comparison Plot

Description

Plots the fitted functions obtained with various regressors (using their default values) on the one-dimensional dataset (X, y).

Usage

comparison_plot(X, y, model_list)

Arguments

X

Predictors

y

Response variables

model_list

List of models to be compared

Examples

sine_data_list <- less::synthetic_sine_curve()
X_sine <- sine_data_list[[1]]
y_sine <- sine_data_list[[2]]

model_list <- c(DecisionTreeRegressor$new(), LinearRegression$new(), KNeighborsRegressor$new())

comparison_plot(X_sine, y_sine, model_list)

Get Functions

Description

Prints the available regressors, clustering methods, tree functions and helper functions within LESS package.

Usage

get_functions()

Examples

get_functions()

k-Fold Cross Validation

Description

Applies k-Fold cross validation to the given model on the given data

Usage

k_fold_cv(
  data = NULL,
  model = NULL,
  random_state = NULL,
  k = 5,
  y_index = ncol(data)
)

Arguments

data

The dataset to be used

model

A classification or a regression model (from LESS package)

random_state

A seed number to get reproducable result

k

Number of splits on the training set (defaults to 5)

y_index

Column index of the response variable on the given data. Default is the last column.

Value

A vector consists of metric of the individual folds and the average metric over the folds

Examples

k_fold_cv(data = iris, model = KNeighborsClassifier$new(), k = 3)

Laplacian kernel - L1 norm

Description

An alternative distance function that can be used in LESS.

Usage

laplacian(data, center, coeff = 0.01)

Arguments

data

Data that includes points in shape of (M x d)

center

A constant point in shape of (1 x d)

coeff

Coefficient value for Laplacian kernel

Value

A numeric vector containing the laplacian kernel distance between each point in data and center.

Examples

data <- matrix(1:12, nrow=3)
center <- c(2, 7, 1, 3)
distances <- laplacian(data, center)
print(distances)

Prepare a Dataset

Description

Takes X and y datasets and merges them into a dataframe with column names (y, X_1, X_2 ...)

Usage

prepareDataset(X, y)

Arguments

X

Independent variables

y

Response variables

Value

A named dataframe which consists of X and y combined

Examples

X <- matrix(1:20, nrow = 4)
y <- c(5:8)
prepareDataset(X, y)

Prepare a Dataset

Description

Takes X dataset and convert it into a dataframe with column names (X_1, X_2 ...)

Usage

prepareXset(X)

Arguments

X

Independent variables

Value

A named dataframe which consists of X

Examples

X <- matrix(1:20, nrow = 4)
prepareXset(X)

RBF kernel - L2 norm

Description

The default distance function in LESS.

Usage

rbf(data, center, coeff = 0.01)

Arguments

data

Data that includes points in shape of (M x d)

center

A constant point in shape of (1 x d)

coeff

Coefficient value for RBF kernel

Value

A numeric vector containing the Radial basis function kernel distance between each point in data and center.

Examples

data <- matrix(1:12, nrow=3)
center <- c(2, 7, 1, 3)
distances <- rbf(data, center)
print(distances)

Synthetic Sine Curve

Description

A simple function to generate n_samples from sine curve in the range (-10, 10) with some amplitude. The function returns the dataset (X, y), and plots the function (curve) along with the dataset (circles)

Usage

synthetic_sine_curve(n_samples = 200)

Arguments

n_samples

Number of data points to be generated

Examples

sine_data_list <- synthetic_sine_curve()
X_sine <- sine_data_list[[1]]
y_sine <- sine_data_list[[2]]

Compare Fitting Time

Description

Plots a histogram chart which shows the fitting time obtained from various regressors/classifiers (using their default values) on the given dataset (X, y).

Usage

test_timing(type = 1, X, y)

Arguments

type

1 to compare regressors, 2 for comparing classifiers

X

Predictors

y

Response variables

Examples

X <- matrix(sample(100, 20), nrow = 10)
y <- sample(100, 10)
test_timing(1, X, y)

Dataset splitting

Description

Split dataframes or matrices into random train and test subsets. Takes the column at the y_index of data as response variable (y) and the rest as the independent variables (X)

Usage

train_test_split(
  data,
  test_size = 0.3,
  random_state = NULL,
  y_index = ncol(data)
)

Arguments

data

Dataset that is going to be split

test_size

Represents the proportion of the dataset to include in the test split. Should be between 0.0 and 1.0 (defaults to 0.3)

random_state

Controls the shuffling applied to the data before applying the split. Pass an int for reproducible output across multiple function calls (defaults to NULL)

y_index

Corresponding column index of the response variable y (defaults to last column of data)

Value

A list of length 4 with elements:

X_train Training input variables
X_test Test input variables
y_train Training response variables
y_test Test response variables

Examples

data(abalone)
split_list <- train_test_split(abalone, test_size =  0.3)
X_train <- split_list[[1]]
X_test <- split_list[[2]]
y_train <- split_list[[3]]
y_test <- split_list[[4]]

print(head(X_train))
print(head(X_test))
print(head(y_train))
print(head(y_test))