datarobot (version 2.8.0)

UploadPredictionDataset: Function to upload new data to a DataRobot project for predictions

Description

The DataRobot prediction engine requires a CSV file containing the data to be used in prediction, and this has been implemented here in two ways. The first and simpler is to specify dataSource as the name of this CSV file, but for the convenience of those who wish to work with dataframes, this function also provides the option of specifying a dataframe, which is then written to a CSV file and uploaded to the DataRobot server.

Usage

UploadPredictionDataset(project, dataSource, forecastPoint = NULL,
  maxWait = 600)

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.

dataSource

object. Either (a) the name of a CSV file (b) a dataframe or (c) url to publicly available file; in each case, this parameter identifies the source of the data for which predictions will be calculated.

forecastPoint

character. Optional. The point relative to which predictions will be generated, based on the forecast window of the project. Only specified in time series projects.

maxWait

integer. The maximum time (in seconds) to wait for each of two steps: (1) The initial dataset upload request, and (2) data processing that occurs after receiving the response to this initial request.

Value

list with the following components:

• id character. The unique alphanumeric identifier for the dataset.

• numColumns numeric. Number of columns in dataset.

• name character. Name of dataset file.

• created character. time of upload.

• projectId character. String giving the unique alphanumeric identifier for the project.

• numRows numeric. Number of rows in dataset.

• forecastPoint character. The point relative to which predictions will be generated, based on the forecast window of the project. Only specified in time series projects, otherwise will be NULL.

Examples

# NOT RUN {
  projectId <- "59a5af20c80891534e3c2bde"
  UploadPredictionDataset(projectId, iris)
# }