SVR: Support Vector Regression

In less: Learning with Subset Stacking

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

• SVR$new()

• SVR$fit()

• SVR$predict()

• SVR$get_estimator_type()

• SVR$clone()

Inherited methods

• less::BaseEstimator$get_all_fields()
• less::BaseEstimator$get_attributes()
• less::BaseEstimator$set_random_state()
• less::SklearnEstimator$get_isFitted()
• less::SklearnEstimator$get_type()

---

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()

less documentation built on Sept. 27, 2022, 5:05 p.m.