list with a single object containing residuals chart data whose name matches the source
requested. See DataPartition for details. This object has the following components:
residualMean. Numeric: the arithmetic mean of the predicted value minus the actual value
over the downsampled dataset.
coefficientOfDetermination. Numeric: aka the r-squared value. This value is calculated
over the downsampled output, not the full input.
data. data.frame: The rows of chart data in [actual, predicted, residual, rowNumber]
form. If the row number was not available at the time of model creation, or if working with
DataRobot 5.2, which does not provide rowNumber in the API response, the rowNumber will be
NA.
histogram. list: Data to plot a histogram of residual values. Each object contains:
intervalStart. Numeric: Start value for an interval, inclusive.
intervalEnd. Numeric: End value for an interval, exclusive for all but the last
interval.
occurrences. Integer: the number of times the predicted value fits within the
interval.
Arguments
model
dataRobotModel. A DataRobot model object like that returned by GetModel. The model
must be a regression model that is not time-aware.
source
character. The data partition for which data would be returned. Default is
DataPartition$VALIDATION. See DataPartition for details.
fallbackToParentInsights
logical. If TRUE, this will return the residuals chart data for
the model's parent if the residuals chart is not available for the model and the model has a
parent model.