K-Fold Split Indices

Problem

Generate k-fold cross-validation splits for n samples (indices 0..n-1).

Return a list of k entries, each of the form [train_indices, val_indices].

Distribute samples so that the first (n % k) folds get one extra sample. Train indices are everything except the validation indices for that fold.

Hint

Slice the indices into k contiguous chunks; train is the complement of val.

case 1

in: [6,3]

expected: [[[2,3,4,5],[0,1]],[[0,1,4,5],[2,3]],[[0,1,2,3],[4,5]]]

case 2

in: [4,2]

expected: [[[2,3],[0,1]],[[0,1],[2,3]]]

case 3

in: [5,5]

expected: [[[1,2,3,4],[0]],[[0,2,3,4],[1]],[[0,1,3,4],[2]],[[0,1,2,4],[3]],[[0,1,2,3],[4]]]

Starter codePython

def kfold_indices(n, k):
    # Returns list of [train_indices, val_indices]
    # Your code here
    pass

3 test cases · run your code against them in the editor