What is k-fold cross-validation?

A. Training k models on the same full dataset B. Using k nearest neighbors only C. Repeatedly training on k-1 folds and validating on the remaining fold D. Testing on the training set k times