Unit test for different subsamples #290
Description
I have noticed that samples generated by the Subsample class all share the same random state.
Line 86 in 2d7fecb
This suggests that all bootstraps are drawned with the same seed. However, quick experiments show that this is not the case.
Would it be possible to add a unit test to ensure that all samples drawn are different ? The same is true for blockbootstrap by the way.
Here is a suggestion :
import numpy as np
import pytest
from mapie.subsample import Subsample
def test_split_SubSample_different_amples() -> None:
"""Test that outputs of subsamplings are all different."""
X = np.array([0, 1, 2, 3])
cv = Subsample(n_resamplings=2, random_state=1)
trains = [x[0] for x in cv.split(X)]
tests = [x[1] for x in cv.split(X)]
with np.testing.assert_raises(AssertionError):
np.testing.assert_equal(trains[0], trains[1])
with np.testing.assert_raises(AssertionError):
np.testing.assert_equal(tests[0], tests[1])