qcut fails with labels=True

[RoyalTS]

The docs say that qcut() takes either a list or a boolean as its labels argument. The argument is None by default. This, which I would have expected to give me the standard set of labels -- like None -- errors:

pd.qcut(range(5), 4, labels=True)

This may well be expected behavior, but it does strike me as semantically weird.

[WillAyd]

The code where this is happening is here:

pandas/pandas/core/reshape/tile.py

Line 353 in a14874f

if labels is not False:

I don't have a strong preference but probably better to support True rather than document as only accepting False (I'm not aware anywhere else where we would do this). If you can come up with a good way to make it work would certainly take a PR

[Dharni0607]

I would like to work on this. I think replacing default value of labels to True would help in resolving the issue.

[jreback]

In [1]: pd.qcut(range(5), 4)                                                                                                                                                                                                                                     
Out[1]: 
[(-0.001, 1.0], (-0.001, 1.0], (1.0, 2.0], (2.0, 3.0], (3.0, 4.0]]
Categories (4, interval[float64]): [(-0.001, 1.0] < (1.0, 2.0] < (2.0, 3.0] < (3.0, 4.0]]

In [2]: pd.cut?                                                                                                                                                                                                                                                  

In [3]: pd.qcut(range(5), 4, labels=list('ABCD'))                                                                                                                                                                                                                
Out[3]: 
[A, A, B, C, D]
Categories (4, object): [A < B < C < D]

In [4]: pd.qcut(range(5), 4, labels=True)                                                                                                                                                                                                                        
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-4-d9fa92b127ec> in <module>
----> 1 pd.qcut(range(5), 4, labels=True)

~/pandas/pandas/core/reshape/tile.py in qcut(x, q, labels, retbins, precision, duplicates)
    312     fac, bins = _bins_to_cuts(x, bins, labels=labels,
    313                               precision=precision, include_lowest=True,
--> 314                               dtype=dtype, duplicates=duplicates)
    315 
    316     return _postprocess_for_cut(fac, bins, retbins, x_is_series,

~/pandas/pandas/core/reshape/tile.py in _bins_to_cuts(x, bins, right, labels, precision, include_lowest, dtype, duplicates)
    357                                     dtype=dtype)
    358         else:
--> 359             if len(labels) != len(bins) - 1:
    360                 raise ValueError('Bin labels must be one fewer than '
    361                                  'the number of bin edges')

TypeError: object of type 'bool' has no len()

In [5]: pd.qcut(range(5), 4, labels=False)                                                                                                                                                                                                                       
Out[5]: array([0, 0, 1, 2, 3])

so the labels=False is giving an indexer and are not 'labels' per se. What would you expect labels=True to actually do?