ComputeUncertainty

is an option for ClassifierMeasurements, LearnedDistribution and other functions to specify if numeric results should be returned along with their uncertainty.

Details

Uncertainties are given using Around.
The uncertainty interval generated typically corresponds to one standard deviation.

Create and test a classifier using ClassifierMeasurements:

Measure the accuracy along with its uncertainty:

Measure the F₁ scores along with their uncertainties:

Train a "Multinormal" distribution on a nominal dataset:

Because of the necessary preprocessing, the PDF computation is not exact:

Use ComputeUncertainty to obtain the uncertainty on the result:

Increase MaxIterations to improve the estimation precision:

Top