cuml_predict: Make predictions on new data points.
In cuml: R Interface for the RAPIDS cuML Suite of Libraries

Use a trained CuML model to make predictions on new data points. Notice calling cuml_predict() will be identical to calling the predict() S3 generic, except for cuml_predict() also comes with proper documentation on all possible predict options (such as output_class_probabilities) and will emit a sensible warning message when a predict option is not applicable for a given model.

1	cuml_predict(model, x, output_class_probabilities = NULL, ...)

`model`	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.
`...`	Additional arguments to `predict()`. Currently unused.