#Python-istas: What libraries do you use for NA-imputation?
I know of, like, none that work. Only, median/mode "imputers" in sklearn.
-
-
Replying to @generativist
@thatdnaguy you want fancier than that? Pandas has some nice stuff, but not much fancier. Is often vectorized.3 replies 1 retweet 1 like -
and, writing your own fancier vectorized one in the pandas context is pretty easy. See blogs of
@TomAugspurger2 replies 0 retweets 2 likes -
IIRC there’s an open PR for scikit-learn with much fancier NA-imputation. Can’t find it ATM
2 replies 0 retweets 0 likes -
Last time I did a Pandas -> R Bridge -> Mice thing. It was unusable.
2 replies 1 retweet 1 like -
at least that sort of thing is easier with feather
1 reply 1 retweet 2 likes -
Ah, great point. I finally have a feather use case!
1 reply 1 retweet 1 like
Replying to @Collinsjo12 @hoffsbeefs and
na translation and other shit if i remember
4:54 PM - 14 Jul 2016
1 reply
0 retweets
1 like
-
-
Replying to @Collinsjo12 @hoffsbeefs and
yea i saw the updates in the mailing list. Very cool
0 replies 0 retweets 0 likes
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.