[Met_help] [rt.rap.ucar.edu #78114] History for how to generate the point_stat output in ascii format

John Halley Gotway via RT met_help at ucar.edu
Thu Sep 29 14:03:30 MDT 2016


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

Dear Mr. or Mrs.,
>From your MET manual, the user can optionally have the duplicated
point_stat output in ascii format.
Would you plz tell me how to turn on this option?
Also, if you have any example scripts to read those output in ascii format
for plotting and can point me to them, that will very helpful.
Thank you.
Ting


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

Subject: how to generate the point_stat output in ascii format
From: John Halley Gotway
Time: Thu Sep 29 10:10:54 2016

Hi Ting,

This is John Halley Gotway.  I see you have a question about
duplicated
point_stat output in ascii format.  I'm not exactly clear what you're
asking, but I have a guess.

In the Point-Stat configuration file, the "output_flaeg" entry
controls the
type of statistics you'd like computed.  There is only setting for
each
line type.  For example, the "cnt" flag controls the output of the CNT
line
type which contains continuous statistics, such as mean error (ME) and
pearson correlation (PR_CORR).  Each flag may to set to one of 3
values:
 - NONE means you don't want that output
 - STAT means you want that output written to the output .stat file.
 - BOTH means you want it written the output .stat file *AND* a
separate
ascii file containing only that line type.

Every time you run Point-Stat, you get an output ascii file with the
suffix
".stat".  That STAT file contains all the computed statistics grouped
together.  So there are continuous statistics lines (CNT) mixed with
categorical counts (CTC) and statistics (CTS), and so on.  Grouping
them
all together makes it difficult for a human to read.  If for example,
you
set "cnt = BOTH", then you'll also get an output file with the suffix
"_cnt.txt" and that file contains only the continuous statistics line
type
and includes a nice header.

Now, how do you plot the output?

  - One option is to write a script using your favorite scripting
language
to create your own plots.

  - Another option is to start with a sample plotting script from the
MET
website:
     http://www.dtcenter.org/met/users/downloads/analysis_scripts.php
     The plot_cnt.R script creates very simple time series plots of
the CNT
line type.
     The plot_mpr.R script creates some summary plots of the MPR line
type.

  - Lastly, there's a whole database and display package for the
output of
MET, called METViewer.  Unfortunately, we don't have funding to
support it
to the broad community.  But here's the web interface for the instance
running at NCAR:
    http://www.dtcenter.org/met/metviewer/metviewer1.jsp

   METViewer is very powerful, but is also somewhat complicated.  I
see you
work at NOAA.  Do you work at NCEP?  There is an instance of METViewer
running there.  You may be able to load your MET output into that
instance
and use it to visualize the output of MET.

Thanks,
John



On Wed, Sep 28, 2016 at 1:54 PM, Ting Lei - NOAA Affiliate via RT <
met_help at ucar.edu> wrote:

>
> Wed Sep 28 13:54:02 2016: Request 78114 was acted upon.
> Transaction: Ticket created by ting.lei at noaa.gov
>        Queue: met_help
>      Subject: how to generate the point_stat output in ascii format
>        Owner: Nobody
>   Requestors: ting.lei at noaa.gov
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=78114 >
>
>
> Dear Mr. or Mrs.,
> From your MET manual, the user can optionally have the duplicated
> point_stat output in ascii format.
> Would you plz tell me how to turn on this option?
> Also, if you have any example scripts to read those output in ascii
format
> for plotting and can point me to them, that will very helpful.
> Thank you.
> Ting
>
>

------------------------------------------------
Subject: how to generate the point_stat output in ascii format
From: Ting Lei - NOAA Affiliate
Time: Thu Sep 29 10:42:25 2016

John,
Your answer is perfectly for me to understand the output.

My original question is because, from the MET user guide,
(Sec. 7.3.3) , it is stated" point_stat poroduces output in STAT and
,optionally, ACII".
Hence, I thought what I need is the second one and thought the STAT is
special format.
So, I will start from your answer along with the MET manual to do the
plot.

Yes, I'm working at EMC/NCEP.  Unfortunately, in addition to the su
permission required,  one colleague told me
METVIEW needs " too many clicks" :).  So, for being now, the most
simple
way for me is to use ncl to plot the STAT.
Maybe I can get some help from NCL help staff.
I think if METviewer is also released that each user can have their
own
copy and control, that will be easier for me to use.

I 'm looking forward to seeing more and more progresses on how EMC
users
use MET when I know there are constant communication between MET and
EMC.


Thank you so much.
Ting







On Thu, Sep 29, 2016 at 12:10 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:

> Hi Ting,
>
> This is John Halley Gotway.  I see you have a question about
duplicated
> point_stat output in ascii format.  I'm not exactly clear what
you're
> asking, but I have a guess.
>
> In the Point-Stat configuration file, the "output_flaeg" entry
controls the
> type of statistics you'd like computed.  There is only setting for
each
> line type.  For example, the "cnt" flag controls the output of the
CNT line
> type which contains continuous statistics, such as mean error (ME)
and
> pearson correlation (PR_CORR).  Each flag may to set to one of 3
values:
>  - NONE means you don't want that output
>  - STAT means you want that output written to the output .stat file.
>  - BOTH means you want it written the output .stat file *AND* a
separate
> ascii file containing only that line type.
>
> Every time you run Point-Stat, you get an output ascii file with the
suffix
> ".stat".  That STAT file contains all the computed statistics
grouped
> together.  So there are continuous statistics lines (CNT) mixed with
> categorical counts (CTC) and statistics (CTS), and so on.  Grouping
them
> all together makes it difficult for a human to read.  If for
example, you
> set "cnt = BOTH", then you'll also get an output file with the
suffix
> "_cnt.txt" and that file contains only the continuous statistics
line type
> and includes a nice header.
>
> Now, how do you plot the output?
>
>   - One option is to write a script using your favorite scripting
language
> to create your own plots.
>
>   - Another option is to start with a sample plotting script from
the MET
> website:
>
http://www.dtcenter.org/met/users/downloads/analysis_scripts.php
>      The plot_cnt.R script creates very simple time series plots of
the CNT
> line type.
>      The plot_mpr.R script creates some summary plots of the MPR
line type.
>
>   - Lastly, there's a whole database and display package for the
output of
> MET, called METViewer.  Unfortunately, we don't have funding to
support it
> to the broad community.  But here's the web interface for the
instance
> running at NCAR:
>     http://www.dtcenter.org/met/metviewer/metviewer1.jsp
>
>    METViewer is very powerful, but is also somewhat complicated.  I
see you
> work at NOAA.  Do you work at NCEP?  There is an instance of
METViewer
> running there.  You may be able to load your MET output into that
instance
> and use it to visualize the output of MET.
>
> Thanks,
> John
>
>
>
> On Wed, Sep 28, 2016 at 1:54 PM, Ting Lei - NOAA Affiliate via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Sep 28 13:54:02 2016: Request 78114 was acted upon.
> > Transaction: Ticket created by ting.lei at noaa.gov
> >        Queue: met_help
> >      Subject: how to generate the point_stat output in ascii
format
> >        Owner: Nobody
> >   Requestors: ting.lei at noaa.gov
> >       Status: new
> >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=78114 >
> >
> >
> > Dear Mr. or Mrs.,
> > From your MET manual, the user can optionally have the duplicated
> > point_stat output in ascii format.
> > Would you plz tell me how to turn on this option?
> > Also, if you have any example scripts to read those output in
ascii
> format
> > for plotting and can point me to them, that will very helpful.
> > Thank you.
> > Ting
> >
> >
>
>

------------------------------------------------
Subject: how to generate the point_stat output in ascii format
From: John Halley Gotway
Time: Thu Sep 29 12:38:23 2016

Ting,

Thanks for the feedback on METViewer.  I can tell you that within the
DTC,
the vast majority of our analysis is done using METViewer.  It is very
powerful, but there's no doubt there's a learning curve.  We clearly
have a
lot more training and documentation to do.

I'll go ahead and resolve this ticket, but let us know if you have
more
questions.

John

On Thu, Sep 29, 2016 at 10:42 AM, Ting Lei - NOAA Affiliate via RT <
met_help at ucar.edu> wrote:

>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=78114 >
>
> John,
> Your answer is perfectly for me to understand the output.
>
> My original question is because, from the MET user guide,
> (Sec. 7.3.3) , it is stated" point_stat poroduces output in STAT and
> ,optionally, ACII".
> Hence, I thought what I need is the second one and thought the STAT
is
> special format.
> So, I will start from your answer along with the MET manual to do
the plot.
>
> Yes, I'm working at EMC/NCEP.  Unfortunately, in addition to the su
> permission required,  one colleague told me
> METVIEW needs " too many clicks" :).  So, for being now, the most
simple
> way for me is to use ncl to plot the STAT.
> Maybe I can get some help from NCL help staff.
> I think if METviewer is also released that each user can have their
own
> copy and control, that will be easier for me to use.
>
> I 'm looking forward to seeing more and more progresses on how EMC
users
> use MET when I know there are constant communication between MET and
EMC.
>
>
> Thank you so much.
> Ting
>
>
>
>
>
>
>
> On Thu, Sep 29, 2016 at 12:10 PM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Hi Ting,
> >
> > This is John Halley Gotway.  I see you have a question about
duplicated
> > point_stat output in ascii format.  I'm not exactly clear what
you're
> > asking, but I have a guess.
> >
> > In the Point-Stat configuration file, the "output_flaeg" entry
controls
> the
> > type of statistics you'd like computed.  There is only setting for
each
> > line type.  For example, the "cnt" flag controls the output of the
CNT
> line
> > type which contains continuous statistics, such as mean error (ME)
and
> > pearson correlation (PR_CORR).  Each flag may to set to one of 3
values:
> >  - NONE means you don't want that output
> >  - STAT means you want that output written to the output .stat
file.
> >  - BOTH means you want it written the output .stat file *AND* a
separate
> > ascii file containing only that line type.
> >
> > Every time you run Point-Stat, you get an output ascii file with
the
> suffix
> > ".stat".  That STAT file contains all the computed statistics
grouped
> > together.  So there are continuous statistics lines (CNT) mixed
with
> > categorical counts (CTC) and statistics (CTS), and so on.
Grouping them
> > all together makes it difficult for a human to read.  If for
example, you
> > set "cnt = BOTH", then you'll also get an output file with the
suffix
> > "_cnt.txt" and that file contains only the continuous statistics
line
> type
> > and includes a nice header.
> >
> > Now, how do you plot the output?
> >
> >   - One option is to write a script using your favorite scripting
> language
> > to create your own plots.
> >
> >   - Another option is to start with a sample plotting script from
the MET
> > website:
> >
http://www.dtcenter.org/met/users/downloads/analysis_scripts.php
> >      The plot_cnt.R script creates very simple time series plots
of the
> CNT
> > line type.
> >      The plot_mpr.R script creates some summary plots of the MPR
line
> type.
> >
> >   - Lastly, there's a whole database and display package for the
output
> of
> > MET, called METViewer.  Unfortunately, we don't have funding to
support
> it
> > to the broad community.  But here's the web interface for the
instance
> > running at NCAR:
> >     http://www.dtcenter.org/met/metviewer/metviewer1.jsp
> >
> >    METViewer is very powerful, but is also somewhat complicated.
I see
> you
> > work at NOAA.  Do you work at NCEP?  There is an instance of
METViewer
> > running there.  You may be able to load your MET output into that
> instance
> > and use it to visualize the output of MET.
> >
> > Thanks,
> > John
> >
> >
> >
> > On Wed, Sep 28, 2016 at 1:54 PM, Ting Lei - NOAA Affiliate via RT
<
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Wed Sep 28 13:54:02 2016: Request 78114 was acted upon.
> > > Transaction: Ticket created by ting.lei at noaa.gov
> > >        Queue: met_help
> > >      Subject: how to generate the point_stat output in ascii
format
> > >        Owner: Nobody
> > >   Requestors: ting.lei at noaa.gov
> > >       Status: new
> > >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=78114
> >
> > >
> > >
> > > Dear Mr. or Mrs.,
> > > From your MET manual, the user can optionally have the
duplicated
> > > point_stat output in ascii format.
> > > Would you plz tell me how to turn on this option?
> > > Also, if you have any example scripts to read those output in
ascii
> > format
> > > for plotting and can point me to them, that will very helpful.
> > > Thank you.
> > > Ting
> > >
> > >
> >
> >
>
>

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


More information about the Met_help mailing list