I would like to know if the following assumption is correct: "In the 10-Fold Cross-Validation we use 100% of data set for the training and 100% of data set for the test"

For the training I'm sure we can say yes, but for the test? I think that for each iteration, if we use different 10% part of data for the test, at the end of the iteration we will have used the entire data set as a test set. It is right?

| cite | improve this question | | | | |

In $k$-fold CV, each of the $k$ splits is used as test set once. Thus, each data point is part of a test set once. So if this is what you mean by '100% of data set for the test', then you are right.

| cite | improve this answer | | | | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.