ENH: Quantiles accepts an array #6955
Conversation
| @@ -482,6 +482,8 @@ Enhancements | |||
| - Added ``how`` option to rolling-moment functions to dictate how to handle resampling; :func:``rolling_max`` defaults to max, | |||
| :func:``rolling_min`` defaults to min, and all others default to mean (:issue:`6297`) | |||
| - ``CustomBuisnessMonthBegin`` and ``CustomBusinessMonthEnd`` are now available (:issue:`6866`) | |||
| - :meth:`Series.quantile` and :meth:`DataFrame.quantile` now accept an array of | |||
| quantiles. | |||
There was a problem hiding this comment.
i think the last line has to be indented
| q : quantile, default 0.5 (50% quantile) | ||
| 0 <= q <= 1 | ||
| q : float or array-like, default 0.5 (50% quantile) | ||
| 0 <= q <= 1, the quantile to compute |
There was a problem hiding this comment.
maybe quantile -> quantile(s)?
|
Thanks, updated. |
| @@ -10945,6 +10945,16 @@ def test_quantile(self): | |||
| xp = df.median() | |||
| assert_series_equal(rs, xp) | |||
|
|
|||
| def test_quantile_multi(self): | |||
| df = DataFrame(np.arange(1, 11).repeat(2).reshape(10, 2), | |||
There was a problem hiding this comment.
maybe test with an empty frame too
|
Added a test on empty frames Also fixed a bug, I wasn't passing the axis argument correctly when quantiles was an array. |
|
One more thing, thinking ahead to fixing #4196; Are we ok with returning a |
|
for |
|
OK. Good to merge? |
|
can you add a test with datetimes both for Series and Frame for completeness? |
|
Do you mean with a DatetimeIndex or Datetime values? It doesn't look like we support quantile on this is off master: In [15]: df
Out[15]:
a b
0 2010-01-01 1
1 2011-01-01 2
2 2012-01-01 3
[3 rows x 2 columns]
In [16]: df.quantile(.5)
Out[16]:
b 2
dtype: float64
In [17]: df.set_index('a').quantile(.5)
Out[17]:
b 2
dtype: float64 |
|
have to pass I think it will blow up actually |
|
iin docstring can you change ref from scipy.score..... to np.percentile (as tht is what it uses now) |
update docs
|
yeah it blows up pretty badly. My simple fix didn't work. Separate issue? |
|
sure |
ENH: Quantiles accepts an array
|
OK. I'll make an issue for that. This PR doesn't officially close any. |
Doesn't quite finish #4196, but it should be easy not that
quantiletakes arrays.I did this on top of #6953, so that should go in first.