Train a spatial prediction and/or interpolation model using Ensemble Machine Learning from a regression/classification matrix
train.spLearner.matrix(
observations,
formulaString,
covariates,
SL.library,
family = stats::gaussian(),
method = "stack.cv",
predict.type,
super.learner,
subsets = 5,
lambda = 0.5,
cov.model = "exponential",
subsample = 10000,
parallel = "multicore",
cell.size,
id = NULL,
weights = NULL,
quantreg = TRUE,
...
)Data frame regression matrix,
Model formula,
SpatialPixelsDataFrame object,
List of learners,
Family e.g. gaussian(),
Ensemble stacking method (see makeStackedLearner),
Prediction type 'prob' or 'response',
Ensemble stacking model usually regr.lm,
Number of subsets for repeated CV,
Target variable transformation for geoR (0.5 or 1),
Covariance model for variogram fitting,
For large datasets consider random subsetting training data,
Initiate parellel processing,
Block size for spatial Cross-validation,
Id column name to control clusters of data,
Optional weights (per row) that learners will use to account for variable data quality,
Fit additional ranger model as meta-learner to allow for derivation of prediction intervals,
other arguments that can be passed on to mlr::makeStackedLearner,
Object of class spLearner