FIX use selected_obj rather the obj throughout groupby #6570
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Ah, something subtle is that df.groupby(as_index=False)[col] is a DataFrameGroupby, hence the need for |
|
This should pass now, will add test iterating over those in 5264 and see what happens. |
| # TODO check groupby with > 1 col ? | ||
|
|
||
| methods = ['count', 'corr', 'cummax', 'cummin', | ||
| 'cumprod', 'describe', 'rank'] |
There was a problem hiding this comment.
fixes for these (at least)
|
Passes travis, and I think this probably fixes for resample and maybe a few others too. Need to add some tests later. Can't get vbench working on this machine, perf change shouldn't be too bad... |
31 tasks
|
@jreback time's look ok... (tests not that consistent), what do you think? |
|
these r fine |
|
I've included the other tests, as I realised that quite a few other functions use groupyby under the hood, quite a bit of fluctuation here... |
|
maybe repeat and see if last 3 show up |
|
If you would, that'd be great! |
|
Here's a patch to fix the perf issue; its that When I run this (even w/o this patch), their are some failing tests....? |
|
Thanks, will apply this, and see what stuff is failing tonight... (weird travis wouldn't be picking them up :s ). Will also add some more basic tests for some of the other methods in the checklist. |
|
I had to rebase off master |
|
@jreback @TomAugspurger Do you mind taking a look at this PR, I see you put some serious time in before trying to sort out this behaviour so would be great to hear what I'm missing (sorry if you were part way there / w duplicated effort here, I'd missed your previous discussion in #5264 until just now). (I had example here with corr, but think it's not an issue) |
|
travis is being flaky. |
|
rebase and I think this is good to go, pls squash a bit too I assume release notes you will do all at once (ok..) |
TST dont ignore subselection in groupby
|
reordered commits and squished (and passes travis). Will add to release when we close main issue. |
hayd
added a commit
that referenced
this pull request
Mar 9, 2014
FIX use selected_obj rather the obj throughout groupby
gouthambs
pushed a commit
to gouthambs/pandas
that referenced
this pull request
Mar 12, 2014
maurosilber
added a commit
to maurosilber/pandas
that referenced
this pull request
Dec 17, 2021
…#44821) Fixes issue pandas-dev#44821. When trying to iterate on a subset of columns in a GroupBy object, it returned all columns, instead of the selected subset. GroupBy.__iter__ used self.obj instead of self._selected_obj (see PR pandas-dev#6570).
4 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
fixes many parts of for #5264... Changes lots of
self.objtoself._selected_obj.probably should vbench before merging?
cc @jreback @TomAugspurger