[MRG+1] Fix bug in StratifiedShuffleSplit for multi-label data with targets having > 1000 labels

[cbrummitt]

This PR fixes a bug for multi-label targets in StratifiedShuffleSplit. The solution being used now is the "label powerset" method: each sequence of labels is mapped to a string with str(row), which transforms a multi-label problem into a multi-class problem.

To see the source of the problem, note that len(str(np.arange(1000))) returns 4056 while len(str(np.arange(1001))) returns 36. The reason is that arrays with > 1000 elements are truncated with an ellipsis: str(np.arange(1001)) gives '[ 0 1 2 ..., 998 999 1000]'. Thus, for multi-label targets with > 1000 labels, samples are mapped onto the same short string whenever their first three values and last three values are the same, which is not the intended behavior.

The solution proposed here, discussed with @vene in this comment thread, is to use ' '.join(row.astype('str')) to convert each target to a string. We are guaranteed that we can do call .astype('str') on row because y = check_array(y, ensure_2d=False, dtype=None) converts y to a numpy array.

As an added benefit, this approach ends up being several faster than str(row) when len(row) < 1000:

In [1]: import numpy as np
In [2]: row = np.random.randint(0, 2, size=500)
In [3]: %timeit ' '.join(row.astype('str'))
169 µs ± 2.39 µs per loop (mean ± std. dev. of 7 runs, 10000 loops each)
In [4]: %timeit str(row)
738 µs ± 23.1 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

I also considered using sklearn.utils.murmurhash3_32 (suggested by @vene as another option) but concluded it was tricky to coerce all the kinds of labels people might use into an int32 data type.

[vene]

Awesome, thank you for catching this and solving it!

Could you also modify the relevant test so that it fails without the patch? Thanks!

[lesteve]

It would be nice to add a test.

[cbrummitt]

Good idea @vene and @lesteve. I added a test for a y with > 1000 labels. I simply added a new function test_stratified_shuffle_split_multilabel_many_labels to test_split.py that fails on the old method and passes with this bug fix.

Is there anything else that would need to be done to hook up this test?

[jnothman]

LGTM

[lesteve]

LGTM, merging, thanks a lot!