📏 Weighted Averaging

Weighted Averaging#

Averaging per-class metrics weighted by class support.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

Weighted averaging is macro averaging with a twist — it computes each class’s metric, then averages them weighted by support (the number of true instances in each class). Larger classes therefore count more.

Its behavior#

This makes the score reflect the actual class distribution — it is essentially macro averaging adjusted for imbalance, sitting between micro and macro. On a representative test set, the weighted average estimates what you’d see on a random production example.

When to use it#

Weighted averaging suits stakeholder and production reporting, where you want one number that respects the real class mix without letting a tiny class swing the result. Use macro instead when minority classes must be weighted equally regardless of frequency.

Theme: Classification & Averaging Metrics · All terminology

Hint

Mind map — connected ideas

Micro Averaging · Macro Averaging · F1-score · Multiclass Classification · Precision (a.k.a. Positive Predictive Value, PPV) · Macro AUC

Hint

More in Classification & Averaging Metrics

Accuracy · AUC (Area Under the Curve) · Average Precision (AP) · Binary Classification · Classification Probability · Discriminatory Power · F1-score · Gini Coefficient · Harmonic Mean · Log Loss (also called Logarithmic Loss or Cross-Entropy Loss) · Macro AUC · Macro AUROC (Macro-Averaged AUROC) · Macro Averaging · Macro F1