aggmetrics: Aggregating Classification Performance Measures for Machine Learning

Ronaldo Cristiano Prati
10/21/2026
cs.LGcs.AI

Official code: rcprati/aggmetrics (0 stars, BSD-3-Clause, last commit 2026-10-23).

Abstract

aggmetrics is a Python package that extends any binary performance measure expressed in the four confusion-matrix counts to multiclass, multilabel, ordinal, cost-sensitive, soft-label and cross-validation settings, by substituting one of four aggregation operators for what libraries usually expose as an average switch. The package also stores, for every measure, the structural properties that decide which identities apply to it, can infer those properties from a user-supplied formula, and uses them to warn when a measure is used where it does not behave as commonly assumed. Its test suite encodes the results of the companion theory as executable tests. The package is BSD-3 licensed, depends only on NumPy, and is available at https://github.com/rcprati/aggmetrics; documentation at https://aggmetrics.readthedocs.io.

View on arXivView PDF

Code Implementations(1)

rcprati/aggmetricsOfficial100%
BSD-3-Clause

Aggregating classification performance measures for machine learning: indicator matrices and aggregation operators

00PythonOct 21, 2026yesterdayBSD-3-Clause
class-imbalanceclassificationevaluation-metricsmachine-learningmultilabel-classification+3 more

Similar Papers

Testing the Consistency of Performance Scores Reported for Binary Classification Problems

73% match1 repo122023

Calibrate: Interactive Analysis of Probabilistic Model Output

72% match2022

A Coherence-Based Measure of AGI

72% match2025

Cost and Reward Infused Metric Elicitation

72% match2025

The MCC-F1 curve: a performance evaluation technique for binary classification

71% match1 repo42020

When Does Synthetic Data Augmentation Improve Score-Based Imbalanced Classification?

71% match2026

Cite this paper

@article{prati2026aggmetrics,
  title  = {aggmetrics: Aggregating Classification Performance Measures for Machine Learning},
  author = {Ronaldo Cristiano Prati},
  year   = {2026},
  eprint = {2610.30456},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2610.30456v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2610.30456.svg)](https://codeofpaper.com/paper/2610.30456)

Discussion