Message-ID: <894CE513-22CA-48BB-B9C7-A2260ED7027A@comcast.net>
Date: 2013-03-29T20:41:18Z
From: David Winsemius
Subject: Installing views in R2.15.3
In-Reply-To: <5155DB32.5070304@statistik.tu-dortmund.de>
On Mar 29, 2013, at 11:19 AM, Uwe Ligges wrote:
>
>
> On 29.03.2013 19:07, David Winsemius wrote:
>>
>> On Mar 29, 2013, at 10:40 AM, ugulumbes at yahoo.com wrote:
>>
>>> Hi,
>>>
>>> Please what am I doing wrong? I tried installing some views (eg MachineLearning, Multivariate) on R2.15.3, but it keeps on telling me that the package is not available for 2.15.3. Is it true?
>>>
>>
>> You cannot install TaskViews. You (the human) need to read the task view and decide what package you wnat to tell your machine to install.
>
>
> Actually, you can install task views via the ctv package, nevertheless, without any reproducible code nor the actual error message it is really hard to help.
My apologies for the incorrect information. You can install the packages in the TaskViews with ctv package. I can see some problems when I try to install the MachineLearning core packages, but this may be due to the fact that I am "ahead of the curve". My R 3.0.0 beta installation under MacOS 10.6.8 already had most if not all of those packages installed but they were at higher version numbers than the ctv installation was requesting from the binary repository.
Got limited success (two packages installed) but several errors relating to 'missing packages' I think because there was a mismatch for ctv's requests for 3.0.0 beta with this command:
install.packages("ctv")
library(ctv)
install.views("MachineLearning", coreOnly = TRUE)
This did run without error:
install.views("MachineLearning", coreOnly = TRUE, type="source")
( Probably needs the XCode to be properly installed if on a Mac or RTools if on Windows.)
> Best,
> Uwe Ligges
>
David Winsemius
Alameda, CA, USA