[ncl-talk] missing value

[Dennis Shea]

It is arbitrary. Set to whatever you want 

Sent from my iPhone

> On Dec 4, 2018, at 1:45 PM, Debasish Hazra <debasish.hazra5 at gmail.com> wrote:
> 
> Thanks Dennis. Is that eps_rain 1e-6 is based on something? or I can change to something bigger like 1e-3 to have less missing values while calculating RMSE ?
> Debasish
> 
>> On Tue, Dec 4, 2018 at 3:20 PM Dennis Shea <shea at ucar.edu> wrote:
>>   eps_rain = 1e-6
>>   sat_rain = where(sat_rain.lt.eps_rain, sat_rain at _FillValue, sat_rain)
>> or
>>   sat_rain = where(sat_rain.lt.eps_rain, 0.0 , sat_rain)
>> 
>>> On Tue, Dec 4, 2018 at 9:37 AM Debasish Hazra <debasish.hazra5 at gmail.com> wrote:
>>> Hi,
>>> I am not able to fix missing value as it seems to be incorrect.
>>> 
>>>          printVarSummary(sat_rain)
>>>           printMinMax(sat_rain,0)
>>>  which shows :
>>> Variable: sat_rain
>>> Type: float
>>> Total Size: 506661120 bytes
>>>             126665280 values
>>> Number of Dimensions: 3
>>> Dimensions and sizes:   [time | 122] x [lat | 721] x [lon | 1440]
>>> Coordinates:
>>>             ttime: [20150601..20150930]
>>>             lat: [ -90..  90]
>>>             lon: [-180..179.75]
>>> Number Of Attributes: 13
>>>   time :        1496275200
>>>   units :       mm
>>>   long_name :  SNIP
>>> .......
>>>   missing_value :       -9999.9
>>>   _FillValue :  -9999.9
>>> (0)     Daily accumulated precipitation (combined microwave-IR) estimate - Late Run (mm) : min=-1.88301e-12   max=595.349
>>> 
>>> To fix the minumum value I assigned :
>>>           sat_fix=sat_rain
>>>           sat_min = min(sat_rain)
>>>            printVarSummary(sat_fix)
>>>           sat_fix at _FillValue = sat_min
>>>           printVarSummary(sat_fix)
>>>           printMinMax(sat_fix,0)
>>> But, it did not change minimum value much : Daily accumulated precipitation (combined microwave-IR) estimate - Late Run (mm) : min=-1.42511e-12   max=595.349
>>> 
>>>   num_missing_orig = num(ismissing(sat_rain))
>>>   num_missing_fix  = num(ismissing(sat_fix))
>>>   print
>>> 
>>> ======================================================================
>>> (0)     Number of original values that are equal to the missing value as
>>> (0)       defined on the file: 29088262
>>> (0)     Number of fixed values that are missing: 29088263
>>> (0)     ======================================================================
>>> 
>>> So, when I am using that variable against model output for RMSE , I am getting warnings :
>>> warning:dim_rmsd_n: 227187 rightmost sections of one or both of the input arrays contained all missing values.
>>> 
>>> Any help on how to correct that is appreciated. I uploaded one input file at ftp, if that helps.
>>> Thanks
>>> Debasish
>>> 
>>> _______________________________________________
>>> ncl-talk mailing list
>>> ncl-talk at ucar.edu
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20181204/8e717809/attachment.html>