#0059

Deterministic Train-Test Split Indices

MediumML20 min20 XP

Problem

Given n rows, a test fraction, and a random seed, return two lists: train indices and test indices. Shuffle indices using random.Random(seed), put the first test_size indices into test, and the rest into train. test_size is round(n * test_fraction).

Why This Matters

A model score is only meaningful if the split is reproducible and the test set stays separate. This problem focuses on split logic before adding any model.

Function Signature

def train_test_indices(n, test_fraction, seed):

Examples

Example 1

Inputn = 5, test_fraction = 0.4, seed = 7

Outputtwo non-overlapping lists covering [0,1,2,3,4]

The exact order is deterministic because the seed controls the shuffle.

Constraints

Use random.Random(seed), not global random.seed.
Return train first, test second.
Each index from 0 to n-1 should appear exactly once across both lists.

CodePython

def train_test_indices(n, test_fraction, seed):
    """
    Input example: n = 5, test_fraction = 0.4, seed = 7
    Return: two non-overlapping lists covering [0,1,2,3,4]
    Return the value from the function so the tests can compare it. Do not print unless the prompt asks.
    """
    # write your code here
    return [], []

def train_test_indices(n, test_fraction, seed):
    """
    Input example: n = 5, test_fraction = 0.4, seed = 7
    Return: two non-overlapping lists covering [0,1,2,3,4]
    Return the value from the function so the tests can compare it. Do not print unless the prompt asks.
    """
    # write your code here
    return [], []

Visible browser tests run here when available.

Testcases2 visible / 3 hidden categories

Split covers all rows

Input10, 0.3, 42

Expected7 train indices and 3 test indices

round(10 * 0.3) gives three test rows.

Reproducible split

Input8, 0.25, 99

Expectedsame output on repeated calls

The seed should make the split deterministic.

Hidden Test Categories

zero test fractionhalf split with odd ndifferent seeds produce different order