NannyML / The-Little-Book-of-ML-Metrics

The book every data scientist needs on their desk.
https://www.nannyml.com/metrics
200 stars 18 forks source link

MAE — Mean Absolute Percentage Error #1

Closed santiviquez closed 2 days ago

santiviquez commented 2 days ago

I already wrote this metric page. I'm just creating the issue as an example.

Metric's name Mean Absolute Percentage Error

Metric's category Regression

Metrics formula

Screenshot 2024-10-09 at 16 36 39

Describe the metrics use cases, and any relevant references. MAE is one of the most popular regression accuracy metrics. It is calculated as the sum of absolute errors divided by the sample size. It is a scale-dependent accuracy measure which means that it uses the same scale as the data being measured.