AUC: Threshold Independent Performance Measures for Probabilistic Classifiers

Various functions compute the area under the curve for four key classification metrics: AUSEC, AUSPC, AUACC, and AUROC. The package, authored by Michel Ballings and Dirk Van den Poel, includes support for visualization and calculating partial areas. It is hosted on the paperswithcode platform.

Use Cases

• Evaluate classifier performance without setting a threshold based on the provided AUSEC, AUSPC, AUACC, and AUROC functions.
• Visualize classifier performance curves using the included visualization support.
• Calculate partial area under the curve for specific performance ranges based on the described functionality.

Strengths

• Implements four distinct area-under-the-curve metrics for classifier evaluation.
• Includes visualization tools and support for partial areas as described.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Michel Ballings and Dirk Van den Poel