datarobot (version 2.18.6)

RequestTransferableModel: Request creation of a transferable model

Description

Requests generation of an transferable model file for use in an on-premise DataRobot standalone prediction environment. This function can only be used if model export is enabled, and will only be useful if you have an on-premise environment in which to import it.

Usage

RequestTransferableModel(project, modelId, predictionIntervalSize = NULL)

Value

jobId

Arguments

project

character. Either (1) a character string giving the unique alphanumeric identifier for the project, or (2) a list containing the element projectId with this identifier.

modelId

numeric. Unique alphanumeric identifier for the model of interest.

predictionIntervalSize

integer. Optional. Added in 2.19. For supervised time series projects, this is the desired prediction interval size for the exported model. A prediction interval is the range of values DataRobot expects actual values of the target to fall within 0 to 100 (inclusive).

Details

This function does not download the exported file. Use DownloadTransferableModel for that.

See Also

Other Transferable Model functions: DeleteTransferableModel(), DownloadTransferableModel(), GetTransferableModel(), ListTransferableModels(), UpdateTransferableModel(), UploadTransferableModel()

Examples

if (FALSE) {
  projectId <- "59a5af20c80891534e3c2bde"
  modelId <- "5996f820af07fc605e81ead4"
  jobId <- RequestTransferableModel(projectId,
                                    modelId,
                                    50)
  WaitForJobToComplete(projectId, jobId)
  file <- file.path(tempdir(), "model.drmodel")
  DownloadTransferableModel(projectObject, modelId, file)
}