[ncl-talk] Convert Matlab code to NCL

[Dave Allured - NOAA Affiliate]

Oops.  Please ignore my correction.  Dennis's version was correct.  The
stride parameter is different between fortran and NCL.

--Dave


On Fri, Oct 6, 2017 at 11:21 AM, Dave Allured - NOAA Affiliate <
dave.allured at noaa.gov> wrote:

> Correction to Dennis's line 1:
>
> Old:  do m=4,0,1
> New:  do m=4,0,-1
>
> I can't confirm the rest of the code.
>
> --Dave
>
>
> On Fri, Oct 6, 2017 at 11:08 AM, Dennis Shea <shea at ucar.edu> wrote:
>
>> A person sent me the following offline:
>>
>> do m=4,0,1
>>    R(:,:,:,m) = dim_sum_n_Wrap(Q(:,:,:,m::,m::,m::),(/3,4,5/))
>>    if m.lt.4 then
>>      R(:,:,:,m) = R(:,:,:,m) - dim_sum_n_Wrap(R(:,:,:,(m+1)::),3)
>>    end if
>> end do
>>
>>
>> On Thu, Oct 5, 2017 at 5:37 PM, <Arnold.Sullivan at csiro.au> wrote:
>>
>>> Dear all ncl users,
>>>
>>>
>>>
>>> Could someone help me to convert Matlab code using sum to NCL
>>>
>>>
>>>
>>> *R is [time] [lat] [lon] [m1]*
>>>
>>>
>>>
>>> *Q is [time] [lat] [lon] [m1] [m2] [m3]*
>>>
>>>
>>>
>>> *For m=5:-1:1*
>>>
>>> *    R(:,:,:,m) = sum( sum( sum( Q(:,:,:,m:end,m:end,m:end), 4), 5), 6);*
>>>
>>> *    if m < 5*
>>>
>>> *        R(:,:,:,m) = R(:,:,:,m) - sum( R(:,:,:,(m+1):end), 4);*
>>>
>>> *end*
>>>
>>> *end*
>>>
>>>
>>>
>>> Is that the same way to use dim_sum_n_Wrap ?
>>>
>>>
>>>
>>> Regards,
>>>
>>>
>>>
>>> Arnold
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20171006/1a6069fc/attachment.html>