AI engineering glossary · ML fundamentals
What is group k-fold?
Group k-fold (also called Grouped cross-validation): A cross-validation method that keeps all rows belonging to one group, such as a single patient or user, together in the same fold. It stops a model from being tested on people whose other records it already saw during training, which would inflate scores.
Group k-fold: an example
In a dataset with 5 scans per patient, all 5 scans of a patient stay either in training or in testing, never split across both.
Related terms
More ML fundamentals terms
- Holdout method
- Hyperparameter
- Hyperparameter tuning
- Hypothesis space
- Independent and identically distributed
- Inductive bias
- Inference
- Inter-annotator agreement
Where you learn it
Group k-fold is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.