Visualization Functions for SptCV Cstf Methods.

Generic S3 plot() and autoplot() (ggplot2) methods to visualize mlr3 spatiotemporal resampling objects.

# S3 method for ResamplingSptCVCstf
autoplot(
  object,
  task,
  fold_id = NULL,
  plot_as_grid = TRUE,
  train_color = "#0072B5",
  test_color = "#E18727",
  tickformat_date = "%Y-%m",
  nticks_x = 3,
  nticks_y = 3,
  point_size = 3,
  axis_label_fontsize = 11,
  static_image = FALSE,
  show_omitted = FALSE,
  plot3D = NULL,
  ...
)

# S3 method for ResamplingRepeatedSptCVCstf
autoplot(
  object,
  task,
  fold_id = NULL,
  repeats_id = 1,
  plot_as_grid = TRUE,
  train_color = "#0072B5",
  test_color = "#E18727",
  tickformat_date = "%Y-%m",
  nticks_x = 3,
  nticks_y = 3,
  point_size = 3,
  axis_label_fontsize = 11,
  plot3D = NULL,
  ...
)

# S3 method for ResamplingSptCVCstf
plot(x, ...)

# S3 method for ResamplingRepeatedSptCVCstf
plot(x, ...)

Arguments

object	[Resampling] mlr3 spatial resampling object of class ResamplingSptCVCstf or ResamplingRepeatedSptCVCstf.
task	[TaskClassifST]/[TaskRegrST] mlr3 task object.
fold_id	[numeric] Fold IDs to plot.
plot_as_grid	[logical(1)] Should a gridded plot using via patchwork be created? If FALSE a list with of ggplot2 objects is returned. Only applies if a numeric vector is passed to argument fold_id.
train_color	[character(1)] The color to use for the training set observations.
test_color	[character(1)] The color to use for the test set observations.
tickformat_date	[character] Date format for z-axis.
nticks_x	[integer] Number of x axis breaks.
nticks_y	[integer] Number of y axis breaks.
point_size	[numeric] Point size of markers.
axis_label_fontsize	[integer] Font size of axis labels.
static_image	[logical] Whether to create a static image from the plotly plot via plotly::orca(). This requires the orca utility to be available. See https://github.com/plotly/orca for more information. When used, by default a file named plot.png is created in the current working directory.
show_omitted	[logical] Whether to show points not used in train or test set for the current fold.
plot3D	[logical] Whether to create a 2D image via ggplot2 or a 3D plot via plotly.
...	Passed down to plotly::orca(). Only effective when static_image = TRUE.
repeats_id	[numeric] Repetition ID to plot.
x	[Resampling] mlr3 spatial resampling object of class ResamplingSptCVCstf or ResamplingRepeatedSptCVCstf.

Details

This method requires to set argument fold_id and no plot containing all partitions can be created. This is because the method does not make use of all observations but only a subset of them (many observations are left out). Hence, train and test sets of one fold are not re-used in other folds as in other methods and plotting these without a train/test indicator would not make sense.

2D vs 3D plotting

This method has both a 2D and a 3D plotting method. The 2D method returns a ggplot with x and y axes representing the spatial coordinates. The 3D method uses plotly to create an interactive 3D plot. Set plot3D = TRUE to use the 3D method.

Note that spatiotemporal datasets usually suffer from overplotting in 2D mode.

See also

• mlr3book chapter on on "Spatiotemporal Visualization".

• Vignette Spatiotemporal Visualization.

• autoplot.ResamplingSpCVBlock()

• autoplot.ResamplingSpCVBuffer()

• autoplot.ResamplingSpCVCoords()

• autoplot.ResamplingSpCVEnv()

• autoplot.ResamplingCV()

• autoplot.ResamplingSptCVCluto()

Examples

# \donttest{
if (mlr3misc::require_namespaces(c("sf", "plotly"), quietly = TRUE)) {
  library(mlr3)
  library(mlr3spatiotempcv)
  task_st = tsk("cookfarm")
  resampling = rsmp("sptcv_cstf",
    folds = 5, time_var = "Date",
    space_var = "SOURCEID")
  resampling$instantiate(task_st)

  # with both `space_var` and `time_var` (LLTO), the omitted observations per
  # fold can be shown by setting `show_omitted = TRUE`
  autoplot(resampling, task_st, fold_id = 1, show_omitted = TRUE)
}
# }