Skip to content

Conversation

jbrockmendel
Copy link
Member

@jbrockmendel jbrockmendel commented Dec 1, 2020

@simonjayhawkins
Copy link
Member

@jbrockmendel #38148 has been reverted, so can you update this to close #37954 and add the release note in 1.2.0 instead.

@jreback jreback added Indexing Related to indexing on series/frames, not to indexes themselves Performance Memory or execution speed performance labels Dec 2, 2020
@jreback jreback added this to the 1.2 milestone Dec 2, 2020
@jreback
Copy link
Contributor

jreback commented Dec 2, 2020

lgtm. the perf enhancements hold here (I think so, just checking).?

@jbrockmendel
Copy link
Member Author

lgtm. the perf enhancements hold here (I think so, just checking).?

i timed it at slightly faster than the copying version (likely within the margin of error, dont recall)

@jreback
Copy link
Contributor

jreback commented Dec 2, 2020

great. ping on green (not sure what's going on with CI).

@jbrockmendel
Copy link
Member Author

not sure what's going on with CI

AFAICT its just backed up. A lot got merged in a short time!

@simonjayhawkins
Copy link
Member

@jreback green

@jreback jreback merged commit c184fde into pandas-dev:master Dec 2, 2020
@jreback
Copy link
Contributor

jreback commented Dec 2, 2020

thanks @jbrockmendel

I think we going to backport this @simonjayhawkins (not sure if the regression was on 1.1.5)?

@simonjayhawkins
Copy link
Member

no evidence this was a regression. maybe can do some digging post 1.2.0rc0 and maybe backport to 1.1.5 and move release note before 1.2.0

but for now, I would say this is not being backported.

@jbrockmendel jbrockmendel deleted the post-38148 branch December 2, 2020 14:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Indexing Related to indexing on series/frames, not to indexes themselves Performance Memory or execution speed performance
Projects
None yet
Development

Successfully merging this pull request may close these issues.

BUG: df.__setitem__ can be 10x slower than pd.concat(..., axis=1)
3 participants