I am able to import the whole data set into R as well, and then select
the required columns. But what I am looking for is to optimize on the
run time, as I only need a few columns out of 100s.
Something similar to what fread (package data.table) provides while
importing from a CSV file.
Utkarsh S
> On 10 Aug 2017, at 14:34 , Peter Dalgaard wrote:
>
> Incidentally, do teach your mailer to not send plain text. It is not much of
> a problem this time, but HTML mails can become quite unreadable on the list.
>
Gah! A "not" remained after editing. DO send plain text, of course.
--
Peter D
hi, the sas universal viewer might be a free, non-R way to convert a
sas7bdat file to non-proprietary formats, not sure if it's windows-only.
those other formats should be easier to import only a subset of columns
into R..
https://support.sas.com/downloads/browse.htm?fil=&cat=74
On Thu, Aug 10, 2
I had a look at this a while back and it didn't seem to be easy. The path of
least resistance would seem to be to use SAS itself to create a data set with
fewer columns, but of course that requires you to get access to SAS.
Otherwise, I think you'd have to modify sas7bdat::read.sas7bdat to drop
Hello everyone,
I want to import data from huge sas files with 100s of columns. The good
thing is that I am only interested in a few selected columns. Is there any
way to do that without loading the full dataset.
I have tried two functions: (1) read.sas7bdat *[from library 'sas7bdat']*,
and (2) r
5 matches
Mail list logo
