Package 'maize'

Title: Specialty Kernels for SVMs
Description: Bindings for SVMs and special kernels via kernlab, e1701, ebmc, and mildsvm extending the 'parsnip' package. Specifically related to kernels for support vector machines not available in parsnip. Package also includes interfaces for novel SVM preprocessors (recipes) and postprocessors (probably).
Authors: Frank Hull [aut, cre], Max Kuhn [ctb]
Maintainer: Frank Hull <[email protected]>
License: MIT + file LICENSE
Version: 0.0.2.95
Built: 2026-05-22 10:40:21 UTC
Source: https://github.com/frankiethull/maize

Help Index

AdaBoost SVM with Radial Basis Function Kernel

Description

AdaBoost SVM with Radial Basis Function Kernel

Usage

ada_boost_svm_rbf(
  mode = "unknown",
  engine = "ebmc",
  num_learners = NULL,
  imb_ratio = NULL
)

Arguments

mode

classification
engine

ebmc's adam2 which uses e1701's svm
num_learners

how many weak learners should be ensembled via boosting
imb_ratio

major-minor class imbalance ratio

Fit an SVM novelty detection model to estimate an applicability domain.

Description

apd_svm_novel_detection() fits an 'one-svc' novelty detection model.

Usage

apd_svm_novel_detection(x, ...)

## Default S3 method:
apd_svm_novel_detection(x, ...)

## S3 method for class 'data.frame'
apd_svm_novel_detection(x, ...)

## S3 method for class 'matrix'
apd_svm_novel_detection(x, ...)

## S3 method for class 'formula'
apd_svm_novel_detection(formula, data, ...)

## S3 method for class 'recipe'
apd_svm_novel_detection(x, data, ...)

Arguments

x

Depending on the context:

  • A data frame of predictors.

  • A matrix of predictors.

  • A recipe specifying a set of preprocessing steps created from recipes::recipe().

...

Options to pass to kernlab::ksvm(). Options should not include data.
formula

A formula specifying the predictor terms on the right-hand side. No outcome should be specified.
data

When a recipe or formula is used, data is specified as:

  • A data frame containing the predictors.

Value

A apd_svm_novel_detection object.

General Interface for "Support Vector Machine" ARIMA Regression Models

Description

arima_svm_laplace() is a way to generate a specification of a time series model that uses SVMs to improve modeling errors (residuals) on Exogenous Regressors. It works with both "automated" ARIMA (auto.arima) and standard ARIMA (arima). The main algorithms are:

  • Auto ARIMA + SVM Errors (engine = auto_arima_svm_laplace, default)

  • ARIMA + SVM Errors (engine = arima_svm_laplace)

Usage

arima_svm_laplace(
  mode = "regression",
  seasonal_period = NULL,
  non_seasonal_ar = NULL,
  non_seasonal_differences = NULL,
  non_seasonal_ma = NULL,
  seasonal_ar = NULL,
  seasonal_differences = NULL,
  seasonal_ma = NULL,
  cost = NULL,
  margin = NULL,
  laplace_sigma = NULL
)

Arguments

mode

A single character string for the type of model. The only possible value for this model is "regression".
seasonal_period

A seasonal frequency. Uses "auto" by default. A character phrase of "auto" or time-based phrase of "2 weeks" can be used if a date or date-time variable is provided. See Fit Details below.
non_seasonal_ar

The order of the non-seasonal auto-regressive (AR) terms. Often denoted "p" in pdq-notation.
non_seasonal_differences

The order of integration for non-seasonal differencing. Often denoted "d" in pdq-notation.
non_seasonal_ma

The order of the non-seasonal moving average (MA) terms. Often denoted "q" in pdq-notation.
seasonal_ar

The order of the seasonal auto-regressive (SAR) terms. Often denoted "P" in PDQ-notation.
seasonal_differences

The order of integration for seasonal differencing. Often denoted "D" in PDQ-notation.
seasonal_ma

The order of the seasonal moving average (SMA) terms. Often denoted "Q" in PDQ-notation.
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
laplace_sigma

sigma parameter for laplacian
sample_size

number for the number (or proportion) of data that is exposed to the fitting routine.

visualize component x-y feature space

Description

visualize component x-y feature space

Usage

## S3 method for class 'kcor_df'
autoplot(x)

Arguments

x

a kcor_df object

RUS Bagged SVM with Radial Basis Function Kernel

Description

RUS Bagged SVM with Radial Basis Function Kernel

Usage

bag_svm_rbf(
  mode = "unknown",
  engine = "ebmc",
  num_learners = NULL,
  imb_ratio = NULL
)

Arguments

mode

classification
engine

ebmc's ub which uses e1701's svm
num_learners

how many weak learners should be ensembled via bagging
imb_ratio

major-minor class imbalance ratio

Uses a support vector regression model to calibrate numeric predictions

Description

Uses a support vector regression model to calibrate numeric predictions

Usage

cal_estimate_svm(
  .data,
  truth = NULL,
  estimate = dplyr::matches("^.pred$"),
  smooth = TRUE,
  parameters = NULL,
  ...,
  .by = NULL
)

## S3 method for class 'data.frame'
cal_estimate_svm(
  .data,
  truth = NULL,
  estimate = dplyr::matches("^.pred$"),
  smooth = TRUE,
  parameters = NULL,
  ...,
  .by = NULL
)

## S3 method for class 'tune_results'
cal_estimate_svm(
  .data,
  truth = NULL,
  estimate = dplyr::matches("^.pred$"),
  smooth = TRUE,
  parameters = NULL,
  ...
)

## S3 method for class 'grouped_df'
cal_estimate_svm(
  .data,
  truth = NULL,
  estimate = NULL,
  smooth = TRUE,
  parameters = NULL,
  ...
)

Arguments

.data

An ungrouped data.frame object, or tune_results object, that contains a prediction column.
truth

The column identifier for the observed outcome data (that is numeric). This should be an unquoted column name.
estimate

Column identifier for the predicted values
smooth

Applies to the svm models. It switches between a polydotTRUE, and vanilladot FALSE.
parameters

(Optional) An optional tibble of tuning parameter values that can be used to filter the predicted values before processing. Applies only to tune_results objects.
...

Additional arguments passed to the models or routines used to calculate the new predictions.

Details

This function uses existing modeling functions from other packages to create the calibration:

These methods estimate the relationship in the unmodified predicted values and then remove that trend when cal_apply() is invoked.

See Also

https://www.tidymodels.org/learn/models/calibration/,

Synthetic Corn Dataset for Corny Example

Description

Synthetic Corn Dataset for Corny Example

Details

Asked Claude Sonnet for a corn data given the README story problem

Value

corn_data

a tibble

Source

claude-3-5-sonnet-20240620

Examples

data(corn_data)
str(corn_data)

Prediction intervals via conformal inference and QRSVM

Description

To compute quantiles, this function uses Quantile SVM instead of probably's "int_conformal_quantile" QRF or classic quantile regression.

Usage

int_conformal_quantile_svm(object, ...)

## S3 method for class 'workflow'
int_conformal_quantile_svm(object, train_data, cal_data, level = 0.95, ...)

Arguments

object

A fitted workflows::workflow() object.
...

Options to pass to qrsvm::qrsvm()
train_data, cal_data

Data frames with the predictor and outcome data. train_data should be the same data used to produce object and cal_data is used to produce predictions (and residuals). If the workflow used a recipe, these should be the data that were inputs to the recipe (and not the product of a recipe).
level

The confidence level for the intervals.

Details

based on the initial probably implementation with slight modification.

origin: https://github.com/tidymodels/probably/blob/HEAD/R/conformal_infer_quantile.R

for more information, visit: https://probably.tidymodels.org

Value

An object of class "int_conformal_quantile" containing the information to create intervals (which includes object). The predict() method is used to produce the intervals.

Kernel Canonical Correlation Analysis

Description

Computes the canonical correlation analysis in feature space. Kernel Canonical Correlation Analysis (KCCA) is a non-linear extension of CCA. Given two random variables (or datasets), KCCA aims at extracting the information which is shared by the two random variables (or datasets). more information found at kernlab::kcca()

Usage

kcca_correlate(x, y = NULL, kernel = "rbfdot", gamma = 0.1, num_comp = 10, ...)

Arguments

x

variable or dataframe
y

variable or dataframe
kernel

a kernel to use
gamma

regularization parameter
num_comp

number of components
...

pass through args for kcca function

Value

A kernel canonical correlation analysis data frame kcor_df

Laplacian Kernel Quantile Regression

Description

laplacian kernel

Usage

kqr_laplace(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  tau = NULL,
  laplace_sigma = NULL
)

Arguments

mode

regression
engine

kernlab kqr
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
tau

a quantile for the loss function
laplace_sigma

sigma parameter for laplacian

Laplacian Least Squares Support Vector Machine

Description

laplacian kernel

Usage

lssvm_laplace(mode = "unknown", engine = "kernlab", laplace_sigma = NULL)

Arguments

mode

classification
engine

kernlab lssvm
laplace_sigma

sigma parameter for laplacian

Multiple Instance SVMs for Ordinal Outcome Data, One-Vs-All, with Radial Basis Function Kernel

Description

Multiple Instance SVMs for Ordinal Outcome Data, One-Vs-All, with Radial Basis Function Kernel

Usage

misvm_orova_rbf(
  mode = "unknown",
  engine = "ebmc",
  cost = NULL,
  rbf_sigma = NULL
)

Arguments

mode

classification
engine

mildsvm::misvm_orova() which uses e1701's svm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
rbf_sigma

A positive number for radial basis function.

RUSBoost SVM with Radial Basis Function Kernel

Description

RUSBoost SVM with Radial Basis Function Kernel

Usage

rus_boost_svm_rbf(
  mode = "unknown",
  engine = "ebmc",
  num_learners = NULL,
  imb_ratio = NULL
)

Arguments

mode

classification
engine

ebmc's rus which uses e1701's svm
num_learners

how many weak learners should be ensembled via boosting
imb_ratio

major-minor class imbalance ratio

Laplacian Relevance Vector Machine (Experimental RVM)

Description

laplacian kernel

Usage

rvm_laplace(
  mode = "unknown",
  engine = "kernlab",
  alpha = NULL,
  var = NULL,
  laplace_sigma = NULL
)

Arguments

mode

regression only for RVM
engine

kernlab rvm
alpha

(alpha) The initial alpha value or vector. Can be either a vector of length equal to the number of data points or a single number.
var

(var) the initial noise variance
laplace_sigma

sigma parameter for laplacian

Predict from a apd_svm_novel_detection

Description

Predict from a apd_svm_novel_detection

Usage

## S3 method for class 'apd_svm_novel_detection'
score(object, new_data, type = "numeric", ...)

Arguments

object

A apd_svm_novel_detection object.
new_data

A data frame or matrix of new samples.
type

A single character. The type of predictions to generate. Valid options are:

  • "numeric" for numeric predictions.

...

Not used, but required for extensibility.

Details

About the score

Value

A tibble of predictions. The number of rows in the tibble is guaranteed to be the same as the number of rows in new_data. The score column is the raw prediction from kernlab::predict() while score_pctl compares this value to the reference distribution of the score created by predicting the training set. A value of X means that X percent of the training data have scores less than the predicted value.

See Also

apd_svm_novel_detection()

Laplacian function kernel KFA signal extraction

Description

step_kfa_laplace() creates a specification of a recipe step that will convert numeric data into one or more kernel components using a laplace kernel. similar to KPCA, but instead of extracting eigenvectors of the dataset in feature space, it approximates the eigenvectors by selecting patterns which are good basis vectors for the dataset.

Usage

step_kfa_laplace(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  num_comp = 5,
  res = NULL,
  columns = NULL,
  sigma = 0.2,
  prefix = "kFA",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kfa_laplace")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
num_comp

The number of components to retain as new predictors. If num_comp is greater than the number of columns or the number of possible components, a smaller value will be used. If num_comp = 0 is set then no transformation is done and selected variables will stay unchanged, regardless of the value of keep_original_cols.
res

An S4 kernlab::kfa() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
sigma

A numeric value for the laplace function parameter.
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfm_nystrom(), step_kha_laplace(), step_kha_tanh(), step_kpca_laplace(), step_kpca_tanh()

Nystrom kernel feature map approximation (RBF)

Description

step_kfm_nystrom() creates a specification of a recipe step that will convert numeric data into a feature appoximation. nystrom approximates the 'radial' kernel approximation.

Usage

step_kfm_nystrom(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  res = NULL,
  columns = NULL,
  sigma = 0.2,
  m = NULL,
  r = NULL,
  sampling = "random",
  prefix = "kFM",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kfm_nystrom")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
res

An mildsvm::kfm_nystrom() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
sigma

A numeric value for the nystrom function parameter.
m

The number rows from df to sample in fitting. defaults to nrow of data
r

The rank of matrix approximation to use. Must be less than or equal to m, the default.
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfa_laplace(), step_kha_laplace(), step_kha_tanh(), step_kpca_laplace(), step_kpca_tanh()

Laplacian function kernel PCA signal extraction via Hebbian Algorithm

Description

step_kha_laplace() creates a specification of a recipe step that will convert numeric data into one or more principal components using a laplace kernel basis expansion.

Usage

step_kha_laplace(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  num_comp = 5,
  res = NULL,
  columns = NULL,
  sigma = 0.2,
  learn_rate = 0.005,
  threshold = 1e-04,
  stop_iter = 100,
  prefix = "kha",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kha_laplace")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
num_comp

The number of components to retain as new predictors. If num_comp is greater than the number of columns or the number of possible components, a smaller value will be used. If num_comp = 0 is set then no transformation is done and selected variables will stay unchanged, regardless of the value of keep_original_cols.
res

An S4 kernlab::kha() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
sigma

A numeric value for the laplace function parameter.
learn_rate

hebbian learning rate
threshold

the smallest value of the convergence step
stop_iter

maximum number of iterations
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfa_laplace(), step_kfm_nystrom(), step_kha_tanh(), step_kpca_laplace(), step_kpca_tanh()

tanh function kernel PCA signal extraction via Hebbian Algorithm

Description

step_kha_tanh() creates a specification of a recipe step that will convert numeric data into one or more principal components using a tanh kernel basis expansion.

Usage

step_kha_tanh(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  num_comp = 5,
  res = NULL,
  columns = NULL,
  scale_factor = 0.2,
  learn_rate = 0.005,
  threshold = 1e-06,
  stop_iter = 100,
  prefix = "kha",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kha_tanh")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
num_comp

The number of components to retain as new predictors. If num_comp is greater than the number of columns or the number of possible components, a smaller value will be used. If num_comp = 0 is set then no transformation is done and selected variables will stay unchanged, regardless of the value of keep_original_cols.
res

An S4 kernlab::kha() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
scale_factor

A numeric value for the tanh function parameter.
learn_rate

hebbian learning rate
threshold

the smallest value of the convergence step
stop_iter

maximum number of iterations
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfa_laplace(), step_kfm_nystrom(), step_kha_laplace(), step_kpca_laplace(), step_kpca_tanh()

Laplacian function kernel PCA signal extraction

Description

step_kpca_laplace() creates a specification of a recipe step that will convert numeric data into one or more principal components using a laplace kernel

Usage

step_kpca_laplace(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  num_comp = 5,
  res = NULL,
  columns = NULL,
  sigma = 0.2,
  prefix = "kPC",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kpca_laplace")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
num_comp

The number of components to retain as new predictors. If num_comp is greater than the number of columns or the number of possible components, a smaller value will be used. If num_comp = 0 is set then no transformation is done and selected variables will stay unchanged, regardless of the value of keep_original_cols.
res

An S4 kernlab::kpca() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
sigma

A numeric value for the laplace function parameter.
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfa_laplace(), step_kfm_nystrom(), step_kha_laplace(), step_kha_tanh(), step_kpca_tanh()

Laplacian function kernel PCA signal extraction

Description

step_kpca_tanh() creates a specification of a recipe step that will convert numeric data into one or more principal components using a tanh kernel basis expansion.

Usage

step_kpca_tanh(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  num_comp = 5,
  res = NULL,
  columns = NULL,
  scale_factor = 0.2,
  offset = 0,
  prefix = "kPC",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("kpca_tanh")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.
...

One or more selector functions to choose variables for this step. See selections() for more details.
role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
trained

A logical to indicate if the quantities for preprocessing have been estimated.
num_comp

The number of components to retain as new predictors. If num_comp is greater than the number of columns or the number of possible components, a smaller value will be used. If num_comp = 0 is set then no transformation is done and selected variables will stay unchanged, regardless of the value of keep_original_cols.
res

An S4 kernlab::kpca() object is stored here once this preprocessing step has be trained by prep().
columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.
scale_factor

A numeric value for the tanh function parameter.
prefix

A character string for the prefix of the resulting new variables. See notes below.
keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.
skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.
id

A character string that is unique to this step to identify it.

See Also

Other multivariate transformation steps: step_kfa_laplace(), step_kfm_nystrom(), step_kha_laplace(), step_kha_tanh(), step_kpca_laplace()

ANOVA RBF Support Vector Machine

Description

anova rbf for support vector machines

Usage

svm_anova_rbf(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  anova_rbf_sigma = NULL,
  degree = NULL,
  margin = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
anova_rbf_sigma

sigma parameter for anova rbf
degree

degree parameter for anova rbf
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Bessel Support Vector Machine

Description

bessel kernel for support vector machines

Usage

svm_bessel(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  bessel_sigma = NULL,
  degree = NULL,
  order = NULL,
  margin = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
bessel_sigma

sigma parameter for bessel
degree

degree parameter for bessel
order

order parameter for bessel
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Cauchy Support Vector Machine

Description

cauchy kernel for support vector machines

Usage

svm_cauchy(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  margin = NULL,
  sigma = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
sigma

a sigma parameter for cauchy kernels

Cosine Similarity Support Vector Machine

Description

cossim kernel for support vector machines

Usage

svm_cossim(mode = "unknown", engine = "kernlab", cost = NULL, margin = NULL)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Fourier Support Vector Machine

Description

fourier kernel for support vector machines

Usage

svm_fourier(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  margin = NULL,
  sigma = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
sigma

a sigma parameter for fourier kernels

Laplacian Support Vector Machine

Description

laplacian kernel for support vector machines

Usage

svm_laplace(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  margin = NULL,
  laplace_sigma = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
laplace_sigma

sigma parameter for laplacian

Sorensen Support Vector Machine

Description

sorensen kernel for support vector machines which is used as a graph kernel for chemical informatics

Usage

svm_sorensen(mode = "unknown", engine = "kernlab", cost = NULL, margin = NULL)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Spline Support Vector Machine

Description

spline kernel for support vector machines

Usage

svm_spline(mode = "unknown", engine = "kernlab", cost = NULL, margin = NULL)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

String Support Vector Machine

Description

stringdot for support vector machines

Usage

svm_string(
  x,
  y,
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  length = NULL,
  lambda = NULL,
  normalized = TRUE,
  margin = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
length

The length of the substrings considered
lambda

The decay factor
normalized

normalize string kernel values, (default = TRUE)
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Hyperbolic Tangent Support Vector Machine

Description

tanh kernel for support vector machines

Usage

svm_tanh(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  scale_factor = NULL,
  margin = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
scale_factor

scale parameter for tanh
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

Tanimoto Support Vector Machine

Description

tanimoto kernel for support vector machines which is used as a graph kernel for chemical informatics

Usage

svm_tanimoto(mode = "unknown", engine = "kernlab", cost = NULL, margin = NULL)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)

T-Student Support Vector Machine

Description

t-student kernel for support vector machines

Usage

svm_tstudent(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  margin = NULL,
  degree = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
degree

a degree parameter for tstudent kernels

Wavelet Support Vector Machine

Description

wavelet kernel for support vector machines

Usage

svm_wavelet(
  mode = "unknown",
  engine = "kernlab",
  cost = NULL,
  margin = NULL,
  sigma = NULL,
  a = 1,
  c = NULL,
  h = NULL
)

Arguments

mode

regression or classification
engine

kernlab ksvm
cost

A positive number for the cost of predicting a sample within or on the wrong side of the margin
margin

A positive number for the epsilon in the SVM insensitive loss function (regression only)
sigma

sigma parameter for svm wavelet kernel
a

scale adjustment parameter for wavelet kernels (temp name)
c

dist adjustment parameter for wavelet kernels can be NULL (temp name)
h

wavelet function for wavelet kernel, default wavelet if NULL (temp name)