torch¶

PyTorch implementation of Metrics.

Note

There is no TorchSingletonCredalSet or TorchDiscreteCredalSet in probly.representation.credal_set.torch; for those semantics, use the numpy-side NumpySingletonCredalSet / NumpyDiscreteCredalSet types. The remaining torch credal sets (Convex, DistanceBased, ProbabilityIntervals, DirichletLevelSet) all use the interval-dominance rule via their lower() / upper() envelopes.

Functions¶

`torch_accuracy`	Compute top-1 classification accuracy for PyTorch tensors.
`torch_auc`	Compute area under a curve using the trapezoid rule.
`torch_average_precision_score`	Compute average precision for PyTorch tensors.
`torch_classwise_ece`	Compute the classwise expected calibration error for PyTorch tensors.
`torch_expected_calibration_error`	Compute the confidence expected calibration error for PyTorch tensors.
`torch_false_negative_rate`	Compute the false negative rate for PyTorch tensors.
`torch_false_positive_rate`	Compute the false positive rate for PyTorch tensors.
`torch_precision_recall_curve`	Compute precision-recall curve along the last axis.
`torch_roc_auc_score`	Compute area under the ROC curve for PyTorch tensors.
`torch_roc_curve`	Compute ROC curve along the last axis.