Creates a complete export bundle containing:
- Vetiver model for R deployment
- ONNX-compatible model files (if supported)
- Preprocessing manifest
- Feature metadata
Usage
export_bundle(
learner,
model_name,
task,
output_dir,
description = NULL,
create_api = FALSE
)
Arguments
- learner
A trained mlr3 Learner or GraphLearner
- model_name
Name for the model bundle
- task
The mlr3 Task used for training
- output_dir
Directory for export files
- description
Optional description
- create_api
Logical, whether to generate Plumber API script
Value
List with paths to all exported files
Examples
if (FALSE) { # \dontrun{
# Create complete export bundle
bundle <- export_bundle(
learner = trained_learner,
model_name = "biomarker_classifier",
task = my_task,
output_dir = "model_export",
create_api = TRUE
)
} # }