Beginner · Data
Class-balanced dataset
Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.
TL;DR. A dataset where each category has roughly an equal number of examples, ensuring fair representation for all classes.
Technical Definition
A dataset where each category has roughly an equal number of examples, ensuring fair representation for all classes.
How it works
A class-balanced dataset is characterized by a relatively even distribution of instances across its different categories or labels. For example, a dataset with 515 instances of native plants and 485 of nonnative plants is considered balanced. This balance is important because significant disparities, often seen in class-imbalanced datasets, can bias a model towards the majority class.
Related Concepts
- Dataset — An organized collection of examples used to train, validate, or test a model.
- Class-imbalanced dataset — A dataset where the number of examples for each class is significantly different, potentially biasing model performance.