represents an object generated by PredictorMeasurements that can be applied to properties.

Details and Options

  • PredictorMeasurementsObject[][prop] is used to look up property prop from the PredictorMeasurementsObject.
  • The following properties can be measured:
  • "BatchEvaluationTime"marginal time to predict one example when a batch is given
    "BestPredictedExamples"examples having the highest actual-class probability density
    "ComparisonPlot"plot of predicted values versus test values
    "EvaluationTime"time needed to predict one example
    "Examples"all test examples
    "Examples"{i1,i2}all examples in the interval i1 predicted in the interval i2
    "GeometricMeanProbabilityDensity"geometric mean of the actual-class probability densities
    "LeastCertainExamples"examples having the highest distribution entropy
    "LogLikelihood"log-likelihood of the model given the test data
    "MeanCrossEntropy"mean cross entropy over test examples
    "MeanDeviation"mean absolute value of the residuals
    "MeanSquare"mean square of the residuals
    "MostCertainExamples"examples having the lowest distribution entropy
    "Perplexity"exponential of the mean cross entropy
    "PredictorFunction"PredictorFunction[] being measured
    "ProbabilityDensities"actual-class prediction probability densities
    "ProbabilityDensityHistogram"histogram of actual-class probability densities
    "Properties"list of measurement properties available
    "RejectionRate"fraction of examples predicted as Indeterminate
    "Report"panel reporting main measurements
    "ResidualHistogram"histogram of residuals
    "ResidualPlot"plot of the residuals
    "Residuals"list of differences between predicted and test values
    "StandardDeviation"root mean square of the residuals
    "StandardDeviationBaseline"standard deviation of test set values
    "WorstPredictedExamples"examples having the lowest actual-class probability density
  • PredictorMeasurementsObject[][prop,opts] specifies that the predictor should use the options opts when applied to the test set. It supersedes options given to PredictorMeasurements.
  • Possible options are as given in PredictorFunction.
  • Test examples classified as Indeterminate are discarded when measuring properties "MeanDeviation", "MeanSquare", "Residuals", "ResidualPlot", "ResidualHistogram" and "StandardDeviation".
  • In properties giving test examples such as "Examples" or "WorstPredictedExamples", examples are in the form inputivaluei where valuei is the actual value.
  • Properties such as "WorstPredictedExamples" or "MostCertainExamples" outputs up to 10 examples. PredictorMeasurementsObject[][propn] can be used to output n example.
  • PredictorMeasurementsObject[] has the same options as PredictorFunction[], with the following additions:
  • WeightsAutomaticweights to be associated with test set examples
    "Uncertainty"Falsewhether measures should be given with their statistical uncertainty
  • With the setting "Uncertainty"True, numerical measures will be given in the form measure±err, where err represents the standard error (corresponding to a 68% confidence interval) associated with measure.
  • Possible settings for weights include:
  • Automaticassociates weight 1 with all test examples
    {w1,w2,}associates weight wi with the i^(th) test examples
  • Changing the weight of a test example from 1 to 2 is equivalent to duplicating this example.
  • Weights affect measures as well as their uncertainties.


Basic Examples  (1)

Define a training set and a test set:

Click for copyable input
Click for copyable input

Create a predictor with the training set:

Click for copyable input

Generate a PredictorMeasurementsObject of the predictor with the test set:

Click for copyable input

Measure the standard deviation and plot the residuals from the PredictorMeasurementsObject:

Click for copyable input

See Also

PredictorMeasurements  PredictorFunction  Predict  PredictorInformation  ClassifierMeasurementsObject

Introduced in 2015
| Updated in 2018