Glossary term
Glossary term
Evaluation and Benchmarks
A mechanism for estimating how well a model would generalize to new data by testing the model against one or more non-overlapping data subsets withheld from the training set.
Created for this library
A pricing analytics team uses 5-fold cross-validation to compare candidate models on small datasets before locking the choice for production.
A medical research team uses leave-one-patient-out cross-validation to avoid optimistic estimates of model accuracy on small clinical datasets.
A risk modeling team uses time-series cross-validation that respects chronological order so the model is never evaluated on data from before its training set.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License