Doc: Adds example of exploding lists into columns instead of storing in dataframe cells #23041
Conversation
…in dataframe cells
Codecov Report
@@ Coverage Diff @@
## master #23041 +/- ##
=======================================
Coverage 92.19% 92.19%
=======================================
Files 169 169
Lines 50873 50873
=======================================
Hits 46904 46904
Misses 3969 3969
Continue to review full report at Codecov.
|
| @@ -336,3 +336,94 @@ constructors using something similar to the following: | |||
| See `the NumPy documentation on byte order | |||
| <https://docs.scipy.org/doc/numpy/user/basics.byteswapping.html>`__ for more | |||
| details. | |||
|
|
|||
|
|
|||
| Alternative to storing lists in DataFrame Cells | |||
There was a problem hiding this comment.
This isn't an "alternative" to lists as much as a way of just reshaping values; this is better phrased as just "Exploding List Items" or something to the effect
|
|
||
| Alternative to storing lists in DataFrame Cells | ||
| ----------------------------------------------- | ||
| Storing nested lists/arrays inside a pandas object should be avoided for performance and memory use reasons. Instead they should be "exploded" into a flat ``DataFrame`` structure. |
There was a problem hiding this comment.
Related to comment above this isn't an "alternative" to using lists within a DataFrame
|
|
||
| .. ipython:: python | ||
|
|
||
| df = pd.DataFrame({'name': ['A.J. Price'] * 3, |
There was a problem hiding this comment.
I believe this is copied from the following SO article:
Need to be careful copy / pasting items from SO into the code base. Would have to get express permission from author to use
There was a problem hiding this comment.
I think SO code snippets are CC BY-SA so as long as we link back to the source (which we should be doing anything) then we're good.
| df | ||
|
|
||
| Stack the new columns as rows; this creates a new index level we'll want to drop in the next step. | ||
| Note that at this point we have a Series, not a Dataframe |
There was a problem hiding this comment.
this needs to be in the cookbook instead. Please cut this down to a much simpler set of examples.
|
What I use to expand a import pandas
genres = pandas.Series([['drama', 'romance'], ['romance'], ['comedy', 'action']])
genres.str.join(',').str.get_dummies(',')So, I think the content of this PR shouldn't be in the documentation (I know you just used the existing PR @mgautam98, sorry I didn't see that earlier). May be a short entry with that to the cookbook could be useful. Closing, if anybody disagrees, please reopen. |
This is the continuation of PR #19215