DOC: Added additional example for groupby by indexer. #13276
Conversation
| @@ -1014,6 +1014,13 @@ Regroup columns of a DataFrame according to their sum, and sum the aggregated on | |||
| df | |||
| df.groupby(df.sum(), axis=1).sum() | |||
|
|
|||
| Groupby by Indexer to 'resample' data. | |||
There was a problem hiding this comment.
This needs a fair bit more explanation of the why and how this does what it does. Maybe show the intent of a resample, then show how one can go about the same idea using non-datetimelike indices.
There was a problem hiding this comment.
need a markdown line here like the other examples
There was a problem hiding this comment.
Can you underline this with ~~~~ to make this a header (see a few lines above this for an example)
There was a problem hiding this comment.
I think you forgot this one
|
Added more documentation and examples to clarify resampling as whole. |
|
|
||
| In order to resample to work on indices that are non-datetimelike , the following procedure can be utilized. | ||
|
|
||
| In the following examples, **df.index / 5** returns a binary array which is used to determine what get's selected for the groupby operation. |
There was a problem hiding this comment.
you can show this (add another ipython block), df.index / 5
|
|
||
| In the following examples, **df.index / 5** returns a binary array which is used to determine what get's selected for the groupby operation. | ||
|
|
||
| .. note:: The above example shows how we can downsample. Downsampling refers to the throwing away of samples. Here by using **df.index / 5**, we are throwing away half the samples. Here we also aggregate samples in bins. By applying **std()** function, we aggregate the information contained in many samples into a small subset of values which is their standard deviation. Hence we can reduce the number of samples by creating bins by clubbing together samples. |
There was a problem hiding this comment.
we don't throw away samples at all. We aggregate them.
|
|
||
| df = pd.DataFrame(np.random.randn(10,2)) | ||
| df | ||
| df.index / 5 |
There was a problem hiding this comment.
This will not work as desired in python 3. Can you make this df.index // 5 to do a floor division explicitly?
|
@pfrcks Added a few more comments. |
|
@jorisvandenbossche kindly go through the changes and comment. |
Current coverage is 84.33%@@ master #13276 diff @@
==========================================
Files 138 138
Lines 50581 51107 +526
Methods 0 0
Messages 0 0
Branches 0 0
==========================================
+ Hits 42582 43103 +521
- Misses 7999 8004 +5
Partials 0 0
|
|
|
||
| In the following examples, **df.index // 5** returns a binary array which is used to determine what get's selected for the groupby operation. | ||
|
|
||
| .. note:: The below example shows how we can downsample which is the throwing away of samples. Here by using **df.index // 5**, we are aggregating the samples in bins. By applying **std()** function, we aggregate the information contained in many samples into a small subset of values which is their standard deviation thereby reducing the number of samples. |
There was a problem hiding this comment.
I still don't like the "throwing away of samples". You don't throw them away, IMO you process them in some way (eg taking the mean of each group of samples)
|
@jorisvandenbossche Sorry, I overlooked that. Have made the necessary changes. Please look and comment. |
|
@pfrcks Thanks! |
git diff upstream/master | flake8 --diff