LAPACK error

[Juan Camilo Plata]

I just installed Revolution R Open. However, when I run a simple regression I got the following error when running summary() to see my regression results. 

Error in chol2inv(Qr$qr[p1, p1, drop = FALSE]) : 

  LAPACK routines cannot be loaded

In addition: Warning message:

In chol2inv(Qr$qr[p1, p1, drop = FALSE]) :

  unable to load shared object 'C:/PROGRA~1/RRO/R-31~1.1/modules/x64/lapack.dll':

  LoadLibrary failure:  The specified procedure could not be found.

Do you have an idea of what is going on here?

Thanks

[David Smith]

Thanks for reporting that, Juan. Not sure what's going on there, but we'll take a look now and get back to you shortly.

Try Revolution Enterprise R Now! 

Get a 14 Day Free Trial of Revolution R Enterprise on AWS Marketplace

[Enzo Martoglio]

Hi
Same here with the notorious R Benchmark 2.5  test by S. Urbanek.

I have loaded RRO on my laptop T410i with i3CPU M330 2.13GHZ - 8GB RAM with Windows 7 SP1

It fails on Laplak segment with error:

  LAPACK routines cannot be loaded
In addition: Warning message:

In eigen(a, symmetric = FALSE, only.values = TRUE) :
  unable to load shared object 'C:/PROGRA~1/RRO/R-3.1.1/

modules/x64/lapack.dll':
  LoadLibrary failure:  The specified procedure could not be found.

As I am not sure where the library is binded by default, I tried as it is and with Revobase installed with no changes.

Thanks Enzo

[David Smith]

Thanks Enzo. We're investigating now to see what the problem is, and will report back here.

# David Smith

[Enzo]

Additional info:  the first time I run the R 35 test (on Sublime Text & R-Box) using Rgui.exe I got an error pop up saying that coudn't find the zgesdd_ module in laplak.dll.
I did not get the same message / pop up with Rstudio (the error message in my previous post has been generated with RSTUDIO). Thanks

[David Smith]

Thanks Enzo, this is helpful info. We'll post back here when we know what's going on.

# David

On Wednesday, October 15, 2014 3:54:48 PM UTC-4, Juan Camilo Plata wrote:

[David Champagne]

To all that have experienced the LAPACK problem on Windows, a new version of Revolution R Open has been posted that fixes the problem

Please perform the following:

1. Uninstall your current installation of Revolution R Open version 8.0

2. Download the new build of Revolution R Open for Windows  (http://mran.revolutionanalytics.com/download/#download)

3. Install the new build

[Juan Camilo Plata]

It is working!!! Thanks!!

[Jonathan Robinson]

Hi, I have installed RRO on a server running CentOS/RedHat (earlier this week in fact). I am getting this same error using the survey package in R. David Smith, you replied that the version for Windows was updated, was the one for Linux also updated?

svy.pew.untrim <- calibrate(svy.unweighted,  bounds=c(0,50),           
+                             formula=~race+income+state+children+married+college+gender+age,
+                             population=c(`(Intercept)`=nrow(popsample) ,
+                                          racetarg, inctarg, statetarg, childtarg, marrtarg, colltarg,
+                                          gendertarg, agetarg    )        )
Error in La.svd(x, nu, nv) : LAPACK routines cannot be loaded

Thanks,

Jonathan R.

[David Champagne]

Hi Johnathan,

I've installed RRO on a RHEL 6.5 machine, and ran the examples for the 'calibrate' function in the survey package documentation, and no errors.  Maybe you should try a re-install

David

Revolution R Plus

Subscribe to Technical Support & Indemnification for R

[David Smith]

Thanks for providing the reproduce, Jonathan. We'll test this out on CentOS and see what we find.

You can track our progress here:

https://github.com/RevolutionAnalytics/RRO/issues/36

Thanks,

# David Smith

-- 

David M Smith <da...@revolutionanalytics.com>
Chief Community Officer, Revolution Analytics  http://blog.revolutionanalytics.com
Tel: +1 (650) 646-9523 (Chicago IL, USA)

Twitter: @revodavid

On Mon, Nov 17, 2014 at 6:31 AM, Jonathan Robinson <jonathanm...@gmail.com> wrote:

library(survey)

data(api)

dclus1<-svydesign(id=~dnum, weights=~pw, data=apiclus1, fpc=~fpc)

pop.totals<-c(`(Intercept)`=6194, stypeH=755, stypeM=1018)

## For a single factor variable this is equivalent to

## postStratify

(dclus1g<-calibrate(dclus1, ~stype, pop.totals))

On Sat, Nov 15, 2014 at 5:32 PM, Jonathan Robinson <jonathanm...@gmail.com> wrote:

I am running CentOS 6.5. I can get you the replicable code using random data on Monday.

On Fri, Nov 14, 2014 at 5:42 PM, David Smith <da...@revolutionanalytics.com> wrote:

Hi Jonathan,

We'll investigate this in more detail. Can you share the specific CentOS (or RedHat) OS version you are using?

If you could also share a call to calibrate that uses built-in data or random data but still shows the error, that would also be helpful.

Thanks,

# David Smith

-- 

David M Smith <da...@revolutionanalytics.com>
Chief Community Officer, Revolution Analytics  http://blog.revolutionanalytics.com
Tel: +1 (650) 646-9523 (Chicago IL, USA)

Twitter: @revodavid

--
You received this message because you are subscribed to the Google Groups "Revolution R Open" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rropen+un...@googlegroups.com.
To post to this group, send email to rro...@googlegroups.com.
Visit this group at http://groups.google.com/group/rropen.

To view this discussion on the web visit https://groups.google.com/d/msgid/rropen/0f8354da-2747-45a4-9110-3f06a60ef747%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Revolution R Plus

Subscribe to Technical Support & Indemnification for R