Re: Re: [ncl-talk] How to get pcvar_varimax when using eofunc_varimax?

Hi Leo,

I would not recommend mixing the depcrated and non-deprecated
functions. If you need to use eofunc_varimax with the pcvar_varimax
attribute, send me a personal email and let me know, and I'll get you
a test binary.

--Mary

On Thu, 19 Apr 2007, LEO_ARIES wrote:

> Hi, Thanks for Mary's help.
>
> I use both "eof_varimax" and "eofunc" at my script, but get a "pcvar_varimax" with all the
> same value. I wonder if a deprecated function and a new feature can be used at the same time.
>
> ; 2. calculate the EOF
> neof = 5
>
> optEOF = True
> optEOF_at_jopt = 1 ; using correlation matrix
> optETS = False
>
> eof = eofunc_Wrap(pr, neof, optEOF)
> eof_ts = eofunc_ts_Wrap (pr, eof, optETS)
> print("Percent Variance of EOF:")
> print(eof_at_pcvar)
>
> ; reof = eofunc_varimax_Wrap(eof,1)
> reof = eof_varimax(eof)
> reof_ts = eofunc_ts_Wrap(pr,reof,optETS)
> print("Percent Variance of REOF:")
> print(reof_at_pcvar_varimax)
>
> And here below is what I get :
> warning:Dimension (0) has not been defined
> (0) Percent Variance of EOF:
> (0) 17.80799
> (1) 12.02987
> (2) 10.94742
> (3) 8.642593
> (4) 7.637198
> warning:Dimension (0) has not been defined
> (0) Percent Variance of REOF:
> (0) 4.545455
> (1) 4.545455
> (2) 4.545455
> (3) 4.545455
> (4) 4.545455
>
> By the way, here is my system infomation :
>
> Linux 2.4.20-8smp #1 SMP Thu Mar 13 17:45:54 EST 2003 i686 i686 i386 GNU/Linux
> gcc (GCC) 3.2.2 20030222 (Red Hat Linux 3.2.2-5)
>
> Thanks
>
>
> Lin
>
>
>
> LEO_ARIES
> 2007-04-19
>
>
>
> ·¢¼þÈË£º Mary Haley
> ·¢ËÍʱ¼ä£º 2007-04-19 03:14:48
> ÊÕ¼þÈË£º LEO_ARIES
> ³­ËÍ£º NCL_Talk; Dennis Shea
> Ö÷Ì⣺ Re: [ncl-talk] How to get pcvar_varimax when using eofunc_varimax?
>
>
>
> Slight correction: you mentioned both the "eof_varimax" and "eofunc_varimax" routines. "eof_varimax" has had the "pcvar_varimax" attribute available since?2005. It is the "eofunc_varimax" routine that had this attributed added for the next scheduled release of NCL.?
>
>
> --Mary
>
>
>
>
>
>
> On Apr 18, 2007, at 10:26 AM, Mary Haley wrote:
>
>
> Hi Lin,
>
>
> The addition of the "pcvar_varimax" attributed was only just recently added, and we failed to indicate this in the documentation. The next release of NCL, which we have slipped a few weeks because of some more GRIB testing, is scheduled for some time around the end of April.
>
>
> If you would like to have a test version, send me an email with information about your system "uname -a". Additionally, if you have a Linux system, I need your "gcc" version as well, "gcc --version".
>
>
> Meanwhile, I will amend the documentation to indicate that "pcvar_varimax" is a new feature.
>
>
> --Mary
> ?
>
> On Apr 18, 2007, at 10:08 AM, LEO_ARIES wrote:
>
>
> Hi,all
> ?
> ??? I've tried to using eofunc_varimax to do a REOF analysis, but I dont know how to get "pcvar_varimax" just
> as said in the introduction of function in NCL page of
> ?
> ?http://www.ncl.ucar.edu/Document/Functions/Built-in/eof_varimax.shtml?
> ?
> ??? Is there anyone who have used this function and tell me how to get this attribute?
> ?
> ?
> Thanks
> ?
> ?
> Lin
> ?
>
>
>
> LEO_ARIES
> 2007-04-19
> _______________________________________________
> ncl-talk mailing list
> ncl-talk_at_ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk_at_ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Thu Apr 19 2007 - 16:06:26 MDT