# API Reference


## Task & Metric Classes


Core classes for defining selective inference tasks and metrics


[SelectiveInferenceTask](SelectiveInferenceTask.md#seapig.SelectiveInferenceTask)  
Wrap a trained `LightningModule` to attach selection results during inference.

[SelectiveMetric](SelectiveMetric.md#seapig.SelectiveMetric)  
Evaluate a metric on full, selected, and rejected subsets.

[RiskCoverageMetric](RiskCoverageMetric.md#seapig.RiskCoverageMetric)  
Build a risk-coverage curve from scores and per-sample errors.

[RiskCoverage](RiskCoverage.md#seapig.RiskCoverage)  
Container for risk-coverage results.


## Embedding based Uncertainty Scores


Uncertainty scoring methods for embedding-based approaches


[scores.CosineScore](scores.CosineScore.md#seapig.scores.CosineScore)  
Returns the KNN-distance based on the cosine distance to the nearest samples.

[scores.EuclideanScore](scores.EuclideanScore.md#seapig.scores.EuclideanScore)  
Returns the KNN-distance based on the Euclidean distance to the nearest samples.

[scores.MahalanobisScore](scores.MahalanobisScore.md#seapig.scores.MahalanobisScore)  
Returns the Mahalanobis distance to the training samples distribution.

[scores.PCAScore](scores.PCAScore.md#seapig.scores.PCAScore)  
Returns uncertainty scores based on PCA reconstruction errors.

[scores.PyODScore](scores.PyODScore.md#seapig.scores.PyODScore)  
Uncertainty scores based on detectors supplied by PyOD.

[scores.utils.TensorPCA](scores.utils.TensorPCA.md#seapig.scores.utils.TensorPCA)  
Tensor-based PCA.


## Logit based Uncertainty Scores


Uncertainty scoring methods based on model logits


[scores.EnergyScore](scores.EnergyScore.md#seapig.scores.EnergyScore)  
Energy-based uncertainty score.

[scores.EntropyScore](scores.EntropyScore.md#seapig.scores.EntropyScore)  
Entropy-based uncertainty score.

[scores.MarginScore](scores.MarginScore.md#seapig.scores.MarginScore)  
Top-two margin uncertainty score.

[scores.SoftmaxScore](scores.SoftmaxScore.md#seapig.scores.SoftmaxScore)  
Maximum softmax probability uncertainty score.


## Other Uncertainty Scores


Uncertainty scoring methods based on other principles


[scores.RandomScore](scores.RandomScore.md#seapig.scores.RandomScore)  
Returns random uncertainty scores per sample.


## Abstract Classes


Abstract base classes


[scores.UncertaintyScore](scores.UncertaintyScore.md#seapig.scores.UncertaintyScore)  
Abstract Base Class for Uncertainty Scores.

[scores.EmbeddingScore](scores.EmbeddingScore.md#seapig.scores.EmbeddingScore)  
Base class for embedding-based uncertainty scores.

[scores.KNNScore](scores.KNNScore.md#seapig.scores.KNNScore)  
Abstract base class for KNN distance-based uncertainty scores.

[scores.LogitScore](scores.LogitScore.md#seapig.scores.LogitScore)  
Base class for logit-based uncertainty scores.


## Utilities


Helpers for progress and logging


[utils.track()](utils.track.md#seapig.utils.track)  
Wrap an iterable with a progress bar.

[utils.enable()](utils.enable.md#seapig.utils.enable)  
Globally enable progress reporting.

[utils.disable()](utils.disable.md#seapig.utils.disable)  
Globally disable progress reporting.

[utils.set_backend()](utils.set_backend.md#seapig.utils.set_backend)  
Select the progress-bar backend.

[utils.get_backend()](utils.get_backend.md#seapig.utils.get_backend)  
Return the name of the currently selected backend.

[utils.get_logger()](utils.get_logger.md#seapig.utils.get_logger)  
Return a logger scoped to the seapig package.

[utils.configure_logging()](utils.configure_logging.md#seapig.utils.configure_logging)  
Configure the seapig package logger.