Seeded Dataset Split Indices

Problem

Split dataset indices into train and validation sets with a seeded shuffle.

Seed the random module, shuffle list(range(n)), then move int(round(n * val_ratio)) indices to validation and keep the rest in train. Return [sorted_train, sorted_val].

Hint

Seed first, shuffle once, then slice the shuffled index list.

Test cases

case 1

in: [5,0.4,42]

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

case 2

in: [6,0.5,1]

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

case 3

in: [1,1,7]

expected: [[],[0]]

case 4

in: [4,0,0]

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

Starter codePython

def split_indices(n, val_ratio, seed):
    # Your code here
    pass

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