Displaying 2 results from an estimated 2 matches for "badx".
Did you mean:
bad
2007 Oct 17
3
Observations on SVD linpack errors, and a workaround
...have before.
From R-Help I see that this issue comes up from time to time.
I'm posting an observation that might help diagnose
the problem, and a workaround that improves the odds of success.
I have found that sometimes svd(t(x)) will work when
svd(x) fails. For example:
> load("badx")
> svd(badx)$d
Error in La.svd(x, nu, nv) : error code 1 from Lapack routine 'dgesdd'
> svd(t(badx))$d
[1] 1.572739e+02 9.614579e+01 7.719867e+01 7.127926e+01 6.490623e+01
.... stuff deleted ....
[126] 8.889272e+00 8.738343e+00 8.447202e+00 8.290393e+00 1.338621e-11
[131] 1...
2009 Feb 10
0
PPR crash (PR#13517)
...t only happens with the inputs are pathological in some way I
don't understand. I have pasted such an example below.
Any help gratefully appreciated.
Regards,
Hugh
### I run the following code to obtain the crash:
bad <- read.csv("C:\\My directories\\bad.csv")
bady <- bad[,3]
badx <- bad[,1:2]
ppr(badx,bady, nterms=1)
### This is the data I read in from the bad.csv file:
"","","bady"
"1",-0.0348207628376986,1,-0.542937988158036
"2",-0.0191633397482220,2,-0.169266091065016
"3",-0.0341944659141195,1,-0.275406839...