cuda.ml (version 0.3.2)

predict.cuda_ml_rand_forest: Make predictions on new data points.

Description

Make predictions on new data points using a CuML random forest model.

Usage

# S3 method for cuda_ml_rand_forest
predict(
  object,
  x,
  output_class_probabilities = NULL,
  cuML_log_level = c("off", "critical", "error", "warn", "info", "debug", "trace"),
  ...
)

Arguments

object

A trained CuML model.

x

A matrix or dataframe containing new data points.

output_class_probabilities

Whether to output class probabilities. NOTE: setting output_class_probabilities to TRUE is only valid when the model being applied is a classification model and supports class probabilities output. CuML classification models supporting class probabilities include knn, fil, and rand_forest. A warning message will be emitted if output_class_probabilities is set to TRUE or FALSE but the model being applied does not support class probabilities output.

cuML_log_level

Log level within cuML library functions. Must be one of "off", "critical", "error", "warn", "info", "debug", "trace". Default: off.

...

Additional arguments to predict(). Currently unused.

Value

Predictions on new data points.