rgpred function

Generate spatial predictions using random forest in ranger (RG)

This function is to make spatial predictions using random forest in ranger.

rgpred(
  trainx,
  trainy,
  longlatpredx,
  predx,
  mtry = if (!is.null(trainy) && !is.factor(trainy)) max(floor(ncol(trainx)/3), 1) else
    floor(sqrt(ncol(trainx))),
  num.trees = 500,
  min.node.size = NULL,
  type = "response",
  num.threads = NULL,
  verbose = FALSE,
  ...
)

Arguments

• trainx: a dataframe or matrix contains columns of predictor variables.
• trainy: a vector of response, must have length equal to the number of rows in trainx.
• longlatpredx: a dataframe contains longitude and latitude of point locations (i.e., the centres of grids) to be predicted.
• predx: a dataframe or matrix contains columns of predictive variables for the grids to be predicted.
• mtry: Number of variables to possibly split at in each node. Default is the (rounded down) square root of the number variables.
• num.trees: number of trees. By default, 500 is used.
• min.node.size: Default 1 for classification, 5 for regression.
• type: Type of prediction. One of 'response', 'se', 'terminalNodes' with default 'response'. See ranger::predict.ranger for details.
• num.threads: number of threads. Default is number of CPUs available.
• verbose: Show computation status and estimated runtime.Default is FALSE.
• ...: other arguments passed on to randomForest.

Returns

A dataframe of longitude, latitude and predictions.

Note

This function is largely based on rfpred.

Examples

## Not run:

data(petrel)
data(petrel.grid)
set.seed(1234)
rgpred1 <- rgpred(petrel[, c(1,2, 6:9)], petrel[, 5], petrel.grid[, c(1,2)],
petrel.grid, num.trees = 500)
names(rgpred1)
## End(Not run)

References

Wright, M. N. & Ziegler, A. (2017). ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R. J Stat Softw 77:1-17. http://dx.doi.org/10.18637/jss.v077.i01.

Author(s)

Jin Li