What is the primary purpose of k-fold cross-validation?
-
A
To obtain a more reliable performance estimate by averaging over multiple train/test splits
-
B
To increase the size of the training set permanently
-
C
To remove outliers from data
-
D
To tune the learning rate automatically