predict.spLearner: Predict using spLearner at new locations

In Envirometrix/landmap: Automated Spatial Prediction using Ensemble Machine Learning

Predict using spLearner at new locations

Description

Predict using spLearner at new locations

Usage

## S3 method for class 'spLearner'
predict(
  object,
  predictionLocations,
  model.error = TRUE,
  error.type = c("forestError", "weighted.sd", "quantreg", "interval")[1],
  t.prob = 1/3,
  w,
  quantiles = c((1 - 0.682)/2, 1 - (1 - 0.682)/2),
  n.cores = parallel::detectCores(),
  what = c("mspe", "bias", "interval"),
  ...
)

Arguments

object	of type spLearner.
predictionLocations	SpatialPixelsDataFrame with values of all features.
model.error	Logical specify if prediction errors should be derived.
error.type	Specify how should be the prediction error be derived.
t.prob	Threshold probability for significant learners; only applyies for meta-learners based on lm model.
w	optional weights vector.
quantiles	Lower and upper quantiles for quantreg forest (0.159 and 0.841 for 1 standard deviation).
n.cores	Number of cores to use (for parallel computation in ranger).
what	A vector of characters indicating what estimates are desired for the quantForestError.
...	optional parameters.

Value

Object of class SpatialPixelsDataFrame with predictions and model error.

Envirometrix/landmap documentation built on June 10, 2022, 10:12 p.m.