k层交叉检验（k-flod cross-validation）

k层交叉检验（k-flod cross-validation）

当然这个称呼也是随着k值的变化而变化的，如果5-flod  cross-validation就叫做5层交叉检验。

 就是说把原始数据分成k份，选一份做为测试数据，剩下的k-1份做为训练数据。

下面附英文解释（To  foreigner  friends）

In K-foldcross-validation, the original sample is randomly partitioned into K subsamples.Of the Ksubsamples, 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 thenrepeated Ktimes (the folds), with each of the K subsamplesused exactly once as the validation data. The K results fromthe folds then can be averaged (or otherwise combined) to produce a singleestimation. The advantage of this method over repeated random sub-sampling isthat all observations are used for both training and validation, and eachobservation is used for validation exactly once. 10-fold cross-validation iscommonly used.