Examples

This is the gallery of examples that showcase the usage of probly. Some examples demonstrate the use of the API in general and some demonstrate specific applications in tutorial form. Also check out our user guide for more detailed illustrations.

Conformal Prediction

Examples concerning the probly.conformal module.

Regression Conformal Prediction — sklearn

Regression Conformal Prediction — sklearn

Classification Conformal Prediction — sklearn

Classification Conformal Prediction — sklearn

Quantile Regression Conformal Prediction — sklearn

Quantile Regression Conformal Prediction — sklearn

Regression Conformal Prediction — PyTorch

Regression Conformal Prediction — PyTorch

Classification Conformal Prediction — PyTorch

Classification Conformal Prediction — PyTorch

Quantile Regression Conformal Prediction — PyTorch

Quantile Regression Conformal Prediction — PyTorch

Plot

Examples concerning the probly.plot module.

Visualising OOD detection results

Visualising OOD detection results

Plotting credal sets on the simplex

Plotting credal sets on the simplex

Plotting binary credal sets on an interval

Plotting binary credal sets on an interval

Plotting credal sets on a spider (radar) chart

Plotting credal sets on a spider (radar) chart

Pytraverse

Examples concerning the pytraverse module.

A Brief Introduction to PyTraverse

A Brief Introduction to PyTraverse

Quantification

Examples concerning the probly.quantification module.

Uncertainty Quantification

Uncertainty Quantification

Release Highlights

These examples illustrate the main features of the releases of probly.

Representation

Examples concerning the probly.representation module.

Singleton credal set

Singleton credal set

Convex credal set

Convex credal set

Discrete credal set

Discrete credal set

Distance-based credal set

Distance-based credal set

Probability-intervals credal set

Probability-intervals credal set

Working with ArraySample

Working with ArraySample

Streaming

Examples that combine probly with online learners on data streams. A single representer() + quantify() call gives you the full aleatoric / epistemic / total decomposition on every step of the stream.

Streaming uncertainty with ARFRegressor

Streaming uncertainty with ARFRegressor

Streaming uncertainty with ARFClassifier

Streaming uncertainty with ARFClassifier

MC-Dropout uncertainty on a 2-D stream

MC-Dropout uncertainty on a 2-D stream

Transformation

Examples concerning the probly.method module.

Applying the Dropout Transformation

Applying the Dropout Transformation

SNGP Distance Awareness on Two Moons

SNGP Distance Awareness on Two Moons