TERMIUM Plus®

The Government of Canada’s terminology and linguistic data bank.

VALIDATION CROISEE K-FOLD [1 record]

Record 1 2022-09-12

English

Subject field(s)
  • Statistical Methods
CONT

In k-fold cross-validation, the original sample is randomly partitioned into k equal sized subsamples. Of the k subsamples, a single subsample is retained as the validation data for testing the model, and the remaining k − 1 subsamples are used as training data. The cross-validation process is then repeated k times, with each of the k subsamples used exactly once as the validation data. The k results can then be averaged to produce a single estimation.

French

Domaine(s)
  • Méthodes statistiques
CONT

La validation croisée à k blocs, «k-fold cross-validation» : on divise l'échantillon original en k échantillons (ou «blocs»), puis on sélectionne un des k échantillons comme ensemble de validation pendant que les k − 1 autres échantillons constituent l'ensemble d'apprentissage. Après apprentissage, on peut calculer une performance de validation. Puis on répète l'opération en sélectionnant un autre échantillon de validation parmi les blocs prédéfinis. À l'issue de la procédure[,] nous obtenons ainsi k scores de performances, un par bloc. La moyenne et l'écart type des k scores de performances peuvent être calculés pour estimer le biais et la variance de la performance de validation.

Spanish

Save record 1

Copyright notice for the TERMIUM Plus® data bank

© Public Services and Procurement Canada, 2024
TERMIUM Plus®, the Government of Canada's terminology and linguistic data bank
A product of the Translation Bureau

Features

Language Portal of Canada

Access a collection of Canadian resources on all aspects of English and French, including quizzes.

Writing tools

The Language Portal’s writing tools have a new look! Easy to consult, they give you access to a wealth of information that will help you write better in English and French.

Glossaries and vocabularies

Access Translation Bureau glossaries and vocabularies.

Date Modified: