Return OOB accuracy

[markgrujic]

Hi,
Thanks for the work on this package. I can't open a branch to submit a PR, but what do you think of the changes here:
https://gist.github.com/markgrujic/a46a4466b618164e77e0abda14d97590

Note a new argument called returnOOB and the associated code from line 80 that returns a list of the imputed data and OOB accuracy if required.
Cheers

[mayer79]

Thx Mark. I will add it very soon. One question: Would it be an option to add the OOB accuracies as an attribute to the data (so that the standard output interface does not change)? I am not yet sure which is more useful.

[markgrujic]

Hi, yep attribute would be a great way to do it too. Either way, I'm not fussed.

[mayer79]

CRAN submission is closed until January 2, but until then you can try to install from github:

library(devtools)
install_github("mayer79/missRanger/release/missRanger")

library(missRanger)
irisWithNA <- generateNA(iris)
irisImputed <- missRanger(irisWithNA, pmm.k = 3, returnOOB = TRUE)
attr(irisImputed, "oob")

#      Species Petal.Length Sepal.Length  Petal.Width  Sepal.Width 
# 0.000000000  0.005397051  0.019491343  0.008372170  0.052749390 

[markgrujic]

Great, works a treat. Sorry for the delay, end of year etc..
Cheers

[mayer79]

Now on CRAN