Skip to content

BUG: groupby.cummin/cummax nans inf values for int64 #15205

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from

Conversation

mroeschke
Copy link
Member

@mroeschke mroeschke commented Jan 24, 2017

I migrated the accum portions of the code into cython and attempted to implement your suggested method to fix this issue @jreback. Let me know if this is along the right track.

@mroeschke mroeschke changed the title BUG: cummin/cummax nans inf values for int64 BUG: groupby.cummin/cummax nans inf values for int64 Jan 24, 2017
@@ -365,7 +365,7 @@ Performance Improvements
- Increased performance of ``pd.factorize()`` by releasing the GIL with ``object`` dtype when inferred as strings (:issue:`14859`)
- Improved performance of timeseries plotting with an irregular DatetimeIndex
(or with ``compat_x=True``) (:issue:`15073`).
- Improved performance of ``groupby().cummin()`` and ``groupby().cummax()`` (:issue:`15048`)
- Improved performance of ``groupby().cummin()`` and ``groupby().cummax()`` (:issue:`15048` & `15109`)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

& `15109` -> , :issue:`151091`

@jreback
Copy link
Contributor

jreback commented Jan 24, 2017

linting issue showing up & small comment, but otherwise lgtm. ping when pushed and green.

@jreback jreback added this to the 0.20.0 milestone Jan 24, 2017
migrate accum into cython

move datetime check into cython

mask int64 inf values before entering cython functions

Edit whatsnew and alter condition

add missing methods

edit space

Linting & whatsnew edit
@codecov-io
Copy link

Current coverage is 85.97% (diff: 100%)

No coverage report found for master at 64d7670.

Powered by Codecov. Last update 64d7670...b390fdb

@mroeschke
Copy link
Member Author

Ping all green.

@jreback
Copy link
Contributor

jreback commented Jan 25, 2017

thanks @mroeschke

@jreback jreback closed this in ba05744 Jan 25, 2017
AnkurDedania pushed a commit to AnkurDedania/pandas that referenced this pull request Mar 21, 2017
closes pandas-dev#15109

Author: Matt Roeschke <[email protected]>

Closes pandas-dev#15205 from mroeschke/fix_15109 and squashes the following commits:

717afb4 [Matt Roeschke] BUG: cummin/cummax nans inf values for int64
@mroeschke mroeschke deleted the fix_15109 branch December 20, 2017 02:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

BUG: groupby.cummin/max changing passed values to nan uncesessarily
3 participants