CV: CV Class

In five-dots/ml4e: Machine Learning for Everyone

Description

A class defines a cross-validation task specification.

Usage

new_cv(options = NULL, keys = NULL, fit_param = NULL)

Arguments

options	Options class object.
keys	A named character vector of keys.
fit_param	A list of fit parameters.

Value

A new CV class object

Super class

ml4e::Task -> CV

Active bindings

fit_param

A list of fit parameters.

fit_call

A fit call.

cv_pred

A data.frame of cross-validation predictions.

test_pred

A data.frame of test predictions.

pred

A data.frame of all predictions.

cv_score

A list of cross-validation score.

test_score

A list of test score.

error_count

A list of error counts against folds

error_count_str

A character of error counts.

done

A logical if finished

Methods

Public methods

• CV$new()

• CV$print()

• CV$get_resample()

• CV$convert_data()

• CV$fit()

• CV$predict()

• CV$get_fitted()

• CV$do()

• CV$clone()

Inherited methods

• ml4e::Task$eval_call_args()

• ml4e::Task$print_task()

---

Method new()

Construct a new CV class object.

Usage

CV$new(options = NULL, keys = NULL, fit_param = NULL)

Arguments

options

A Options class object.

keys

A named character vector of keys.

fit_param

A list of fit parameters. If NULL, keys must have "fit_param" key.

Returns

A new CV class object

---

Method print()

Print object.

Usage

CV$print()

---

Method get_resample()

Get a resample.

Usage

CV$get_resample(seed = NULL)

Arguments

seed

A integer scalar seed.

Returns

A resample of rset class object.

---

Method convert_data()

Convert data by self$data_call.

Usage

CV$convert_data(data)

Arguments

data

A data.frame or rsplit object to be converted.

Returns

A converted data appropreate for the model input.

---

Method fit()

Fit model by self$fit_call.

Usage

CV$fit(data = NULL)

Arguments

data

A data.frame or rsplit object.

Returns

A model fitted object.

---

Method predict()

Predict by a model.

Usage

CV$predict(fitted = NULL, new_data = NULL, ids = NULL)

Arguments

fitted

A model fitted object.

new_data

A data.frame to be predicted.

ids

A integer vector of row indexes.

Returns

A data.frame of predicted values.

---

Method get_fitted()

Get a fitted object by all training data.

Usage

CV$get_fitted()

---

Method do()

Run a cross-validation. The result will be stored in self$result.

Usage

CV$do(cv_seed = TRUE)

Arguments

cv_seed

A logical scalar whether to use self$cv_seed. If FALSE, then self$search_seed is used.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

CV$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

five-dots/ml4e documentation built on June 19, 2020, 4:26 p.m.