Spatial Buffer Cross validation implemented by the blockCV
package.
However, the default settings allow to conduct a leave-one-out cross validation for two-class, multi-class and continuous response data, where each observation is one test set. For each test, all observations outside the buffer around the test observation are included in the training set.
The parameter spDataType = PB
and addBG
are designed for
presence-background data in species distribution modelling. If spDataType = PB
, test sets are only created for each presence observation
(task\$positive
). The option addBG = TRUE
adds the background data inside
the buffer to the corresponding test sets. For each test set, all
observations outside the buffer around the test observation are included in
the training set.
Valavi R, Elith J, Lahoz-Monfort JJ, Guillera-Arroita G (2018). “blockCV: an R package for generating spatially or environmentally separated folds for k-fold cross-validation of species distribution models.” bioRxiv. doi: 10.1101/357798 .
mlr3::Resampling
-> ResamplingSpCVBuffer
iters
integer(1)
Returns the number of resampling iterations, depending on the
values stored in the param_set
.
new()
Create an "Environmental Block" resampling instance.
ResamplingSpCVBuffer$new(id = "spcv_buffer")
id
character(1)
Identifier for the resampling strategy.
instantiate()
Materializes fixed training and test splits for a given task.
ResamplingSpCVBuffer$instantiate(task)
task
Task
A task to instantiate.
clone()
The objects of this class are cloneable with this method.
ResamplingSpCVBuffer$clone(deep = FALSE)
deep
Whether to make a deep clone.
if (mlr3misc::require_namespaces(c("sf", "blockCV"), quietly = TRUE)) { library(mlr3) task = tsk("ecuador") # Instantiate Resampling rcv = rsmp("spcv_buffer", theRange = 10000) rcv$instantiate(task) # Individual sets: rcv$train_set(1) rcv$test_set(1) intersect(rcv$train_set(1), rcv$test_set(1)) # Internal storage: # rcv$instance }#> integer(0)