@raghavrv please check, i was working on machinelearnjs/machinelearnjs#243, found that these tests were written for wrong set of inputs, fixed that.

Sorry, can you please clarify what you're fixing?
@raghavrv is no longer contributing.

@amueller I forgot the exact change but
X = np.arange(7)
y = np.asarray([0, 1, 1, 1, 2, 2, 2])
a = StratifiedShuffleSplit(3, 0.2).split(X, y)

next(a)

we get error:

Traceback (most recent call last):
File "", line 1, in
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1315, in split
for train, test in self._iter_indices(X, y, groups):
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1695, in _iter_indices
raise ValueError("The least populated class in y has only 1"
ValueError: The least populated class in y has only 1 member, which is too few. The minimum number of groups for any class cannot be less than 2.
`

b = StratifiedShuffleSplit(3, 2).split(X, y)

next(b)

we get the same error

Traceback (most recent call last):
File "", line 1, in
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1315, in split
for train, test in self._iter_indices(X, y, groups):
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1695, in _iter_indices
raise ValueError("The least populated class in y has only 1"
ValueError: The least populated class in y has only 1 member, which is too few. The minimum number of groups for any class cannot be less than 2.

but if:
X = np.arange(9)
y = np.asarray([0, 0, 0, 1, 1, 1, 2, 2, 2])
c = StratifiedShuffleSplit(3, 2).split(X, y)
next(c)

we get error

Traceback (most recent call last):
File "", line 1, in
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1315, in split
for train, test in self._iter_indices(X, y, groups):
File "/mnt/code/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_split.py", line 1707, in _iter_indices
(n_test, n_classes))
ValueError: The test_size = 2 should be greater or equal to the number of classes = 3

which is the same error as said in comment # Check that error is raised if the test set size is smaller than n_classes

so I think that should be the order of test cases.
😄

@amueller please review.

There is no longer assert_raises in test_split.py, which has been largely reorganized in the meanwhile.
I am closing this one.