[datatable-help] Fwd: New package dataframe

Tim Hesterberg rocket at google.com
Tue May 22 19:39:20 CEST 2012


Resending, now that I'm on this list.

---------- Forwarded message ----------
From: Tim Hesterberg <rocket at google.com>
Date: Tue, May 22, 2012 at 9:44 AM
Subject: Re: New package dataframe
To: Matthew Dowle <mdowle at mdowle.plus.com>
Cc: datatable-help at lists.r-forge.r-project.org


Two other comments:
* look at dataframe/inst/doc/differences.txt - that describes the
programming tricks I used, you might find them useful
* however, Luke Tierney is looking at that too and trying to change R to
make those tricks unnecessary.  He made a change to the development version
that may make the "attributes<-"(...) trick I use unnecessary.


On Tue, May 22, 2012 at 9:34 AM, Tim Hesterberg <rocket at google.com> wrote:

> Hi Matthew,
>
> I haven't done timings to compare it to data.table.  Chris Neff has done
> some, and I think data.table is faster.
>
> I have done benchmarks for number of copies made.  data.table is usually
> better than dataframe alone,
> and the combination is usually best of all.  I'll attach those.
>
> Tim
>
> For my records:
> ~/R/data.table/memory.R
>
>
> On Tue, May 22, 2012 at 2:45 AM, Matthew Dowle <mdowle at mdowle.plus.com>wrote:
>
>>
>> Hi Tim,
>>
>> http://cran.r-project.org/web/packages/dataframe/index.html
>>
>> Looks interesting. How does it differ to data.table? I had a quick look
>> but couldn't see any timings.
>>
>> Matthew
>>
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/datatable-help/attachments/20120522/b2aa2a4d/attachment.html>


More information about the datatable-help mailing list