ENH + BUG: insert of new values for axis parameter on pandas.DataFrame.quantile #9544
Conversation
|
you don't need all this.
mainly need to add a test |
|
You are right. I am correcting it. |
|
I corrected it and I implemented a basic test. In case tell me if I did any mistake. |
| axis : {0, 1} | ||
| 0 for row-wise, 1 for column-wise | ||
| axis : {0, 1, 'index', 'columns'} (default 0) | ||
| 0 or 'index' a quantile for each column, 1 or 'columns' a |
There was a problem hiding this comment.
I don't think this is as clear as the 0 for row-wise, .... What about 0 or 'index' for row-wise, ...?
There was a problem hiding this comment.
Hi Kevin,
I think you are right. I found it more clear defining exactly what the result will be (because I found that non-native english speakers often struggle to understand the meaning of -wise: see http://forum.wordreference.com/showthread.php?t=403654 ). But I think that your observation is correct.
|
I finally had the time to complete my PR with the Travis CI build. Now what have I to do? (it is my first PR) |
| @@ -11786,6 +11786,30 @@ def test_quantile(self): | |||
| expected = Series([3., 4.], index=[0, 1]) | |||
| assert_series_equal(result, expected) | |||
|
|
|||
| def test_quantile_axis_parameter(self): | |||
| from numpy import percentile | |||
There was a problem hiding this comment.
can you add the issue number here
|
this looks fine. can you add a release note for 0.17.0 in the API section? ping when travis is green. |
|
Hi @jreback
Thanks |
|
@fbrundu this the file to which to add release notes: https://github.com/pydata/pandas/blob/master/doc/source/whatsnew/v0.16.2.txt (we're actually going to release 0.16.2 before 0.17, and this can go in there) As for the issue number, yes that would be fine. Something as simple as |
|
@jreback travis is green but git tells me merge conflicts |
|
merged via c0917de thanks! |
closes #9543
I added the values "index" and "columns" for the axis parameter of the method pandas.DataFrame.quantile().
Also, there was no check on the value inserted for axis. E.g. if 2 or "foo" was inserted, no ValueError() was raised.
I added a basic check.
Thanks