mlr_resamplings_forecast_cv: Forecast Cross-Validation Resampling

In mlr-org/mlr3forecasting: Forecasting for 'mlr3'

Forecast Cross-Validation Resampling

Description

Splits data using a folds-folds (default: 10 folds) rolling window cross-validation.

Dictionary

This Resampling can be instantiated via the dictionary mlr_resamplings or with the associated sugar function rsmp():

mlr_resamplings$get("forecast_cv")
rsmp("forecast_cv")

Parameters

• folds (integer(1))
  Number of folds.

• window_size (integer(1))
  (Minimal) Size of the rolling window.

• horizon (integer(1))
  Forecasting horizon in the test sets.

• fixed_window (logial(1))
  Flag for fixed sized window. If FALSE an expanding window is used.

Super class

mlr3::Resampling -> ResamplingForecastCV

Active bindings

iters

(integer(1))
Returns the number of resampling iterations, depending on the values stored in the param_set.

Methods

Public methods

• ResamplingForecastCV$new()

• ResamplingForecastCV$clone()

Inherited methods

• mlr3::Resampling$format()
• mlr3::Resampling$help()
• mlr3::Resampling$instantiate()
• mlr3::Resampling$print()
• mlr3::Resampling$test_set()
• mlr3::Resampling$train_set()

---

Method new()

Creates a new instance of this R6 class.

Usage

ResamplingForecastCV$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

ResamplingForecastCV$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

References

Bergmeir, Christoph, Hyndman, J R, Koo, Bonsoo (2018). “A note on the validity of cross-validation for evaluating autoregressive time series prediction.” Computational Statistics & Data Analysis, 120, 70–83.

See Also

• as_benchmark_result() to convert to a BenchmarkResult.

• Chapter in the mlr3book: https://mlr3book.mlr-org.com/performance.html#sec-resampling

• Package mlr3viz for some generic visualizations.

Other resample: mlr_resamplings_forecast_holdout

Examples

# Create a task with 10 observations
task = mlr3::tsk("airpassengers")
task$filter(1:20)

# Instantiate Resampling
rfho = mlr3::rsmp("forecast_cv", folds = 3, fixed_window = FALSE)
rfho$instantiate(task)

# Individual sets:
rfho$train_set(1)
rfho$test_set(1)
intersect(rfho$train_set(1), rfho$test_set(1))

# Internal storage:
rfho$instance #  list

mlr-org/mlr3forecasting documentation built on June 29, 2023, 11:57 p.m.