[Met_help] [rt.rap.ucar.edu #92746] History for METviewer batch jobs: when fcst valid time is the independent variable

Tatiana Burek via RT met_help at ucar.edu
Wed Oct 23 07:31:09 MDT 2019


----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

Hello.  I'm trying to do a batch plot of a time series using "forecast 
valid time" as independent variable, and am not sure how to specify the 
independent variables without spelling out all the forecast valid 
times.  In the attached three files, the 1st and 2nd are the *.png and 
*.xml of the interactive METviewer plot - the *.xml spells out the 
Independent variables (a full list of valid times):

         <indep equalize="true" name="fcst_valid_beg">
             <val label="2019-08-01 12:00:00" plot_val="">2019-08-01 
12:00:00</val>
             <val label="2019-08-02 12:00:00" plot_val="">2019-08-02 
12:00:00</val>

                       ....

             <val label="2019-10-13 12:00:00" plot_val="">2019-10-13 
12:00:00</val>
             <val label="2019-10-14 12:00:00" plot_val="">2019-10-14 
12:00:00</val>
         </indep>

How should the above be handled in a batch job?  When the forecast valid 
time is not an independent variable (merely a fixed variable), I'd do

     <date_range name="vtimes">
       <start>2019-08-01 12:00:00</start>
       <end>2019-10-14 12:00:00</end>
       <inc>86400</inc>
       <format>yyyy-MM-dd HH:mm:ss</format>
     </date_range>

then later

         <plot_fix>
             <field equalize="true" name="fcst_valid_beg">
                <date_range name="vtimes"/>
             </field>

             .....

          </plot_fix>

The attached plt_rmse.xml was my first attempt at making the time series 
plot in batch, and the independent variable set up obviously needs more 
work, but I can't figure out how to properly "unroll" the specified 
<date_range name="vtimes"> to create the equivalent of the above 
independent variables list.

Thank you very much -

Ying

-- 
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov




----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: METviewer batch jobs: when fcst valid time is the independent variable
From: Tatiana Burek
Time: Tue Oct 22 13:48:37 2019

Ying,

<date_range> doesn't work for the independent variables. You should
use <date_list> instead.
In plt_rmse.xml:
- remove <date_range...
- instead of <date_range name="vtimes"/> under <indep> put:
          <date_list>
              <start>2019-08-01 12:00:00</start>
              <end>2019-10-14 12:00:00</end>
              <inc>86400</inc>
              <label_format>yyyy-MM-dd HH:mm:ss</label_format>
          </date_list>
Please notice <label_format> and not <format>

Tatiana

On Tue Oct 22 10:55:45 2019, ying.lin at noaa.gov wrote:
> Hello.  I'm trying to do a batch plot of a time series using
"forecast
> valid time" as independent variable, and am not sure how to specify
the
> independent variables without spelling out all the forecast valid
> times.  In the attached three files, the 1st and 2nd are the *.png
and
> *.xml of the interactive METviewer plot - the *.xml spells out the
> Independent variables (a full list of valid times):
>
>          <indep equalize="true" name="fcst_valid_beg">
>              <val label="2019-08-01 12:00:00" plot_val="">2019-08-01
> 12:00:00</val>
>              <val label="2019-08-02 12:00:00" plot_val="">2019-08-02
> 12:00:00</val>
>
>                        ....
>
>              <val label="2019-10-13 12:00:00" plot_val="">2019-10-13
> 12:00:00</val>
>              <val label="2019-10-14 12:00:00" plot_val="">2019-10-14
> 12:00:00</val>
>          </indep>
>
> How should the above be handled in a batch job?  When the forecast
valid
> time is not an independent variable (merely a fixed variable), I'd
do
>
>      <date_range name="vtimes">
>        <start>2019-08-01 12:00:00</start>
>        <end>2019-10-14 12:00:00</end>
>        <inc>86400</inc>
>        <format>yyyy-MM-dd HH:mm:ss</format>
>      </date_range>
>
> then later
>
>          <plot_fix>
>              <field equalize="true" name="fcst_valid_beg">
>                 <date_range name="vtimes"/>
>              </field>
>
>              .....
>
>           </plot_fix>
>
> The attached plt_rmse.xml was my first attempt at making the time
series
> plot in batch, and the independent variable set up obviously needs
more
> work, but I can't figure out how to properly "unroll" the specified
> <date_range name="vtimes"> to create the equivalent of the above
> independent variables list.
>
> Thank you very much -
>
> Ying
>



------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #92746] METviewer batch jobs: when fcst valid time is the independent variable
From: Ying Lin
Time: Tue Oct 22 15:04:35 2019

Hi Tatiana,

     It works, this is great!  Thank you so much for your help!

     Please close this ticket at your convenience.

Ying

On 10/22/19 3:48 PM, Tatiana Burek via RT wrote:
> Ying,
>
> <date_range> doesn't work for the independent variables. You should
use <date_list> instead.
> In plt_rmse.xml:
> - remove <date_range...
> - instead of <date_range name="vtimes"/> under <indep> put:
>            <date_list>
>                <start>2019-08-01 12:00:00</start>
>                <end>2019-10-14 12:00:00</end>
>                <inc>86400</inc>
>                <label_format>yyyy-MM-dd HH:mm:ss</label_format>
>            </date_list>
> Please notice <label_format> and not <format>
>
> Tatiana
>
> On Tue Oct 22 10:55:45 2019, ying.lin at noaa.gov wrote:
>> Hello.  I'm trying to do a batch plot of a time series using
"forecast
>> valid time" as independent variable, and am not sure how to specify
the
>> independent variables without spelling out all the forecast valid
>> times.  In the attached three files, the 1st and 2nd are the *.png
and
>> *.xml of the interactive METviewer plot - the *.xml spells out the
>> Independent variables (a full list of valid times):
>>
>>           <indep equalize="true" name="fcst_valid_beg">
>>               <val label="2019-08-01 12:00:00" plot_val="">2019-08-
01
>> 12:00:00</val>
>>               <val label="2019-08-02 12:00:00" plot_val="">2019-08-
02
>> 12:00:00</val>
>>
>>                         ....
>>
>>               <val label="2019-10-13 12:00:00" plot_val="">2019-10-
13
>> 12:00:00</val>
>>               <val label="2019-10-14 12:00:00" plot_val="">2019-10-
14
>> 12:00:00</val>
>>           </indep>
>>
>> How should the above be handled in a batch job?  When the forecast
valid
>> time is not an independent variable (merely a fixed variable), I'd
do
>>
>>       <date_range name="vtimes">
>>         <start>2019-08-01 12:00:00</start>
>>         <end>2019-10-14 12:00:00</end>
>>         <inc>86400</inc>
>>         <format>yyyy-MM-dd HH:mm:ss</format>
>>       </date_range>
>>
>> then later
>>
>>           <plot_fix>
>>               <field equalize="true" name="fcst_valid_beg">
>>                  <date_range name="vtimes"/>
>>               </field>
>>
>>               .....
>>
>>            </plot_fix>
>>
>> The attached plt_rmse.xml was my first attempt at making the time
series
>> plot in batch, and the independent variable set up obviously needs
more
>> work, but I can't figure out how to properly "unroll" the specified
>> <date_range name="vtimes"> to create the equivalent of the above
>> independent variables list.
>>
>> Thank you very much -
>>
>> Ying
>>
>
>

--
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov



------------------------------------------------


More information about the Met_help mailing list