UPENN | Midterm | K Fold Validation | DA Practice |

Question

Suppose we want to compute $10-Fold$ Cross-Validation error on $100$ training examples. We need to compute error $N1$ times, and the Cross-Validation error is the average of the errors. To compute each error, we need to build a model with data of size $N2$, and test the model on the data of size $N3$. What are the appropriate numbers for $N1, N2, N3$?

 
(a) $N1 = 10, N2 = 90, N3 = 10$

(b) $N1 = 1, N2 = 90, N3 = 10$

(c) $N1 = 10, N2 = 100, N3 = 10$

(d) $N1 = 10, N2 = 100, N3 = 10$

Answer 1

N1 = 10:

• In 10-fold cross-validation, the data is divided into 10 equal folds.
• Each fold is used as a test set once, while the remaining 9 folds are used for training.
• Therefore, we need to compute the error 10 times (once for each fold).

N2 = 90:

• For each of the 10 iterations, we use 90% of the data (90 examples) for training the model.
• This is because we're using 9 out of the 10 folds for training in each iteration.

N3 = 10:

• The remaining 10% of the data (10 examples) are used as the test set in each iteration.
• This is the fold that wasn't used for training in that particular iteration.