[Met_help] [rt.rap.ucar.edu #98957] History for building a scorecard using metviewer gui
John Halley Gotway via RT
met_help at ucar.edu
Thu Mar 18 10:11:57 MDT 2021
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Good morning,
I have a scorecard xml template but would like to first build a scorecard
using the interactive metviewer gui so that I can optimize what I'd like to
put in the scorecard. Can that be done? It's not clear that it can on the
gui. Ideally I'd like to build rows for various met/chemistry fields for
different stats, and from there tabulate the day1, day2, and day3
performance for key regions. Also, the scorecards that I've seen have
red/green indicators with size of the symbols indicating a greater
difference between the models examined. Gray areas are not reported. Is
there a placeholder value that is assigned to areas in the table that do
not report any statistics? thanks.
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: building a scorecard using metviewer gui
From: John Halley Gotway
Time: Mon Mar 08 09:39:07 2021
Ed,
Unfortunately, no, there is no GUI interface for creating a scorecard.
It
must all be done with trial-and-error via the batch engine. If you
think
this would be a very useful enhancement, you could recommend it to
Jason
for inclusion in a future development and funding cycle.
I'm not exactly sure what you mean by a placeholder value. Generally,
the
up/down pointing red/green arrows indicate statistically significant
differences. In some cells, the differences may not be statistically
significant. And that's different from having no data for a cell.
Here's
the current documentation about scorecards:
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
If you have more detailed questions about it, I could refer you to
Tatiana.
Hope that helps.
Thanks,
John
On Mon, Mar 8, 2021 at 4:48 AM Edward Strobach - NOAA Affiliate via RT
<
met_help at ucar.edu> wrote:
>
> Mon Mar 08 04:48:26 2021: Request 98957 was acted upon.
> Transaction: Ticket created by edward.strobach at noaa.gov
> Queue: met_help
> Subject: building a scorecard using metviewer gui
> Owner: Nobody
> Requestors: edward.strobach at noaa.gov
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957 >
>
>
> Good morning,
>
> I have a scorecard xml template but would like to first build a
scorecard
> using the interactive metviewer gui so that I can optimize what I'd
like to
> put in the scorecard. Can that be done? It's not clear that it can
on the
> gui. Ideally I'd like to build rows for various met/chemistry
fields for
> different stats, and from there tabulate the day1, day2, and day3
> performance for key regions. Also, the scorecards that I've seen
have
> red/green indicators with size of the symbols indicating a greater
> difference between the models examined. Gray areas are not
reported. Is
> there a placeholder value that is assigned to areas in the table
that do
> not report any statistics? thanks.
>
>
------------------------------------------------
Subject: building a scorecard using metviewer gui
From: Edward Strobach - NOAA Affiliate
Time: Mon Mar 08 12:51:50 2021
Hi John,
I was having a difficult time finding the right thing to say. I knew
about
the size difference and their meaning, but what if there is nothing to
report on in one of the squares? Does that just get 'grayed' out?
I'll
give trial and error a try. Logan did send an xml template, but I
wanted
to confirm my understanding first. Here's my understanding as far as
I
know it. Note that in Logan's original template he had only CSI and
RMSE
for visibility and CAPE I believe. I'm taking largely from that but
expanding it to fit my needs. Here's a copy of what I had sent if you
have
time to confirm my logic:
Hi Logan,
I think I understand most of the scorecard layout in the xml file.
The template you sent creates rows first followed by columns. You
have
listed CSI, which would span, on the far left column, all the rows
where
thresholds are being reported. The next column over would be the
fields
(second column), followed by thresholds in the third column. You also
have
a second stat variable after CSI called FBIAS, although no variables
appear
to be assigned there. This is what I've done so far:
* <rows> <field name="stat"> <val name="CSI"
label="Critical
Success Index"/> <field name="fcst_var"> <val
name="PMAVE"
label="Daily Ave. PM2.5"/> <field name="obs_thresh">
<val name="<10" label="< 10 ug/m3"/> <val
name="<20"
label="< 20 ug/m3"/> <val name="<30" label="< 30
ug/m3"/> <val name="<40" label="< 40 ug/m3"/>
<val name="<50" label="< 50 ug/m3"/> <val
name="<60"
label="< 60 ug/m3"/> </field> </field>
<field
name="fcst_var"> <val name="PDMAX1" label="Daily Max.
PM2.5"/>
<field name="obs_thresh"> <val name="<10"
label="<
10 ug/m3"/> <val name="<20" label="< 20 ug/m3"/>
<val name="<30" label="< 30 ug/m3"/> <val
name="<40"
label="< 40 ug/m3"/> <val name="<50" label="< 50
ug/m3"/> <val name="<60" label="< 60 ug/m3"/>
</field> </field> <field name="fcst_var"> <val
name="OZMAX8" label="8-hour Max. Ozone"/> <field
name="obs_thresh"> <val name="<50" label="< 50 ppb"/>
<val name="<65" label="< 65 ppb"/> <val
name="<80" label="< 80 ppb"/> <val name="<100"
label="< 100 ppb"/> <val name="<125" label="< 125
ppb"/> <val name="<150" label="< 150 ppb"/>
</field> </field> <field name="fcst_var"> <val
name="OZMAX1" label="Hourly Max. Ozone"/> <field
name="obs_thresh"> <val name="<50" label="< 50 ppb"/>
<val name="<65" label="< 65 ppb"/> <val
name="<80" label="< 80 ppb"/> <val name="<100"
label="< 100 ppb"/> <val name="<125" label="< 125
ppb"/> <val name="<150" label="< 150 ppb"/>
</field>*
* </field> <val name="FBIAS" label="FBIAS"/>
<field
name="fcst_var"> <val name="PMTF" label="PM 2.5"/>
</field> <field name="fcst_var"> <val name="OZCON"
label="Ozone"/> </field> <field name="fcst_var">
<val name="TMP" label="2-m Temp"/> </field> <field
name="fcst_var"> <val name="DPT" label="2-m Dewpoint"/>
</field> <field name="fcst_var"> <val name="WSPD10"
label="10-m Wind Speed"/> </field> <val name="RMSE"
label="RMSE"/> <field name="fcst_var"> <val
name="PMTF"
label="PM 2.5"/> </field> <field name="fcst_var">
<val name="OZCON" label="Ozone"/> </field> <field
name="fcst_var"> <val name="TMP" label="2-m Temp"/>
</field> <field name="fcst_var"> <val name="DPT"
label="2-m
Dewpoint"/> </field> <field name="fcst_var">
<val
name="WSPD10" label="10-m Wind Speed"/> </field> </field>
</rows>*
This would translate to something like this I think, or something
similar
(using the defined "label" of course):
________________________
| | <=10
| | <=20
| | <=30
CSI | PMAVE | <=40
* | | <=50*
* | | <=60*
* |___________________*
* |*
I save the max/ave. fields for CSI and focus on the other fields for
FBIAS
and RMSE. I marked in red what I was surprised by since I thought
that I
might have to close the val name with all fields of interest contained
inside the xml block. In other words, I expected something like this
instead:
*<val name="RMSE" label="RMSE">*
* <field name="fcst_var"> <val name="PMTF" label="PM
2.5"/> </field>*
*....*
*....*
*</val>*
For columns I believe I can do something similar. I can first start
by
region (so have NWC, SWC, SEC, etc.)
Under each of these larger columns would be day1, day2, and day3, i.e.
| NWC |
| day 1 | day 2 | day3 |
In the past I was able to use a value name for forecast hour. In this
case
it seems that I would want to group 0-23, 24-47, and 48-71. I think
this
grouping should be possible but I'm not sure about the appropriate
syntax
to do so. Also, if there's nothing to report, do I just Null it out
in the
xml file or does that happen automatically.
Hopefully I was clear in my email, but if I wasn't, then just let me
know.
Thanks.
On Mon, Mar 8, 2021 at 11:39 AM John Halley Gotway via RT
<met_help at ucar.edu>
wrote:
> Ed,
>
> Unfortunately, no, there is no GUI interface for creating a
scorecard. It
> must all be done with trial-and-error via the batch engine. If you
think
> this would be a very useful enhancement, you could recommend it to
Jason
> for inclusion in a future development and funding cycle.
>
> I'm not exactly sure what you mean by a placeholder value.
Generally, the
> up/down pointing red/green arrows indicate statistically significant
> differences. In some cells, the differences may not be statistically
> significant. And that's different from having no data for a cell.
Here's
> the current documentation about scorecards:
>
>
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
>
> If you have more detailed questions about it, I could refer you to
Tatiana.
>
> Hope that helps.
>
> Thanks,
> John
>
> On Mon, Mar 8, 2021 at 4:48 AM Edward Strobach - NOAA Affiliate via
RT <
> met_help at ucar.edu> wrote:
>
> >
> > Mon Mar 08 04:48:26 2021: Request 98957 was acted upon.
> > Transaction: Ticket created by edward.strobach at noaa.gov
> > Queue: met_help
> > Subject: building a scorecard using metviewer gui
> > Owner: Nobody
> > Requestors: edward.strobach at noaa.gov
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957 >
> >
> >
> > Good morning,
> >
> > I have a scorecard xml template but would like to first build a
scorecard
> > using the interactive metviewer gui so that I can optimize what
I'd like
> to
> > put in the scorecard. Can that be done? It's not clear that it
can on
> the
> > gui. Ideally I'd like to build rows for various met/chemistry
fields for
> > different stats, and from there tabulate the day1, day2, and day3
> > performance for key regions. Also, the scorecards that I've seen
have
> > red/green indicators with size of the symbols indicating a greater
> > difference between the models examined. Gray areas are not
reported. Is
> > there a placeholder value that is assigned to areas in the table
that do
> > not report any statistics? thanks.
> >
> >
>
>
------------------------------------------------
Subject: building a scorecard using metviewer gui
From: John Halley Gotway
Time: Mon Mar 08 15:44:47 2021
Ed (and Tatiana),
I was hoping that the existing documentation would have had an
explanation
of that. But reading through this:
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
I see a reference to this:
*<threshold_file>:* full path to the XML file describing
configurations for
thresholds, colors and symbols (optional).
But there's no further explanation of that threshold_file. Presumably
it'd
contain options to specify how missing data cells are colored in. So
we
need more explanation and examples of that. I've cc'ed Tatiana on this
to
let her know that we're missing documentation/examples of the
threshold_file. I went poking around the METviewer repo but didn't
immediately find any good info about that.
Thanks,
John
On Mon, Mar 8, 2021 at 12:52 PM Edward Strobach - NOAA Affiliate via
RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957 >
>
> Hi John,
>
> I was having a difficult time finding the right thing to say. I
knew about
> the size difference and their meaning, but what if there is nothing
to
> report on in one of the squares? Does that just get 'grayed' out?
I'll
> give trial and error a try. Logan did send an xml template, but I
wanted
> to confirm my understanding first. Here's my understanding as far
as I
> know it. Note that in Logan's original template he had only CSI and
RMSE
> for visibility and CAPE I believe. I'm taking largely from that but
> expanding it to fit my needs. Here's a copy of what I had sent if
you have
> time to confirm my logic:
>
> Hi Logan,
>
> I think I understand most of the scorecard layout in the xml file.
>
> The template you sent creates rows first followed by columns. You
have
> listed CSI, which would span, on the far left column, all the rows
where
> thresholds are being reported. The next column over would be the
fields
> (second column), followed by thresholds in the third column. You
also have
> a second stat variable after CSI called FBIAS, although no variables
appear
> to be assigned there. This is what I've done so far:
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> * <rows> <field name="stat"> <val name="CSI"
label="Critical
> Success Index"/> <field name="fcst_var"> <val
name="PMAVE"
> label="Daily Ave. PM2.5"/> <field name="obs_thresh">
> <val name="<10" label="< 10 ug/m3"/> <val
name="<20"
> label="< 20 ug/m3"/> <val name="<30" label="< 30
> ug/m3"/> <val name="<40" label="< 40 ug/m3"/>
> <val name="<50" label="< 50 ug/m3"/> <val
name="<60"
> label="< 60 ug/m3"/> </field> </field>
<field
> name="fcst_var"> <val name="PDMAX1" label="Daily Max.
PM2.5"/>
> <field name="obs_thresh"> <val name="<10"
label="<
> 10 ug/m3"/> <val name="<20" label="< 20 ug/m3"/>
> <val name="<30" label="< 30 ug/m3"/> <val
name="<40"
> label="< 40 ug/m3"/> <val name="<50" label="< 50
> ug/m3"/> <val name="<60" label="< 60 ug/m3"/>
> </field> </field> <field name="fcst_var">
<val
> name="OZMAX8" label="8-hour Max. Ozone"/> <field
> name="obs_thresh"> <val name="<50" label="< 50
ppb"/>
> <val name="<65" label="< 65 ppb"/> <val
> name="<80" label="< 80 ppb"/> <val name="<100"
> label="< 100 ppb"/> <val name="<125" label="<
125
> ppb"/> <val name="<150" label="< 150 ppb"/>
> </field> </field> <field name="fcst_var">
<val
> name="OZMAX1" label="Hourly Max. Ozone"/> <field
> name="obs_thresh"> <val name="<50" label="< 50
ppb"/>
> <val name="<65" label="< 65 ppb"/> <val
> name="<80" label="< 80 ppb"/> <val name="<100"
> label="< 100 ppb"/> <val name="<125" label="<
125
> ppb"/> <val name="<150" label="< 150 ppb"/>
> </field>*
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> * </field> <val name="FBIAS" label="FBIAS"/>
<field
> name="fcst_var"> <val name="PMTF" label="PM 2.5"/>
> </field> <field name="fcst_var"> <val name="OZCON"
> label="Ozone"/> </field> <field name="fcst_var">
> <val name="TMP" label="2-m Temp"/> </field> <field
> name="fcst_var"> <val name="DPT" label="2-m Dewpoint"/>
> </field> <field name="fcst_var"> <val name="WSPD10"
> label="10-m Wind Speed"/> </field> <val name="RMSE"
> label="RMSE"/> <field name="fcst_var"> <val
name="PMTF"
> label="PM 2.5"/> </field> <field name="fcst_var">
> <val name="OZCON" label="Ozone"/> </field> <field
> name="fcst_var"> <val name="TMP" label="2-m Temp"/>
> </field> <field name="fcst_var"> <val name="DPT"
label="2-m
> Dewpoint"/> </field> <field name="fcst_var">
<val
> name="WSPD10" label="10-m Wind Speed"/> </field>
</field>
> </rows>*
>
> This would translate to something like this I think, or something
similar
> (using the defined "label" of course):
>
> ________________________
> | | <=10
> | | <=20
> | | <=30
> CSI | PMAVE | <=40
> * | | <=50*
>
> * | | <=60*
>
> * |___________________*
>
> * |*
> I save the max/ave. fields for CSI and focus on the other fields for
FBIAS
> and RMSE. I marked in red what I was surprised by since I thought
that I
> might have to close the val name with all fields of interest
contained
> inside the xml block. In other words, I expected something like
this
> instead:
>
> *<val name="RMSE" label="RMSE">*
>
>
> * <field name="fcst_var"> <val name="PMTF" label="PM
> 2.5"/> </field>*
> *....*
> *....*
> *</val>*
>
> For columns I believe I can do something similar. I can first start
by
> region (so have NWC, SWC, SEC, etc.)
> Under each of these larger columns would be day1, day2, and day3,
i.e.
>
>
> | NWC |
> | day 1 | day 2 | day3 |
>
> In the past I was able to use a value name for forecast hour. In
this case
> it seems that I would want to group 0-23, 24-47, and 48-71. I think
this
> grouping should be possible but I'm not sure about the appropriate
syntax
> to do so. Also, if there's nothing to report, do I just Null it out
in the
> xml file or does that happen automatically.
>
> Hopefully I was clear in my email, but if I wasn't, then just let me
know.
> Thanks.
>
> On Mon, Mar 8, 2021 at 11:39 AM John Halley Gotway via RT <
> met_help at ucar.edu>
> wrote:
>
> > Ed,
> >
> > Unfortunately, no, there is no GUI interface for creating a
scorecard. It
> > must all be done with trial-and-error via the batch engine. If you
think
> > this would be a very useful enhancement, you could recommend it to
Jason
> > for inclusion in a future development and funding cycle.
> >
> > I'm not exactly sure what you mean by a placeholder value.
Generally, the
> > up/down pointing red/green arrows indicate statistically
significant
> > differences. In some cells, the differences may not be
statistically
> > significant. And that's different from having no data for a cell.
Here's
> > the current documentation about scorecards:
> >
> >
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
> >
> > If you have more detailed questions about it, I could refer you to
> Tatiana.
> >
> > Hope that helps.
> >
> > Thanks,
> > John
> >
> > On Mon, Mar 8, 2021 at 4:48 AM Edward Strobach - NOAA Affiliate
via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Mon Mar 08 04:48:26 2021: Request 98957 was acted upon.
> > > Transaction: Ticket created by edward.strobach at noaa.gov
> > > Queue: met_help
> > > Subject: building a scorecard using metviewer gui
> > > Owner: Nobody
> > > Requestors: edward.strobach at noaa.gov
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957
> >
> > >
> > >
> > > Good morning,
> > >
> > > I have a scorecard xml template but would like to first build a
> scorecard
> > > using the interactive metviewer gui so that I can optimize what
I'd
> like
> > to
> > > put in the scorecard. Can that be done? It's not clear that it
can on
> > the
> > > gui. Ideally I'd like to build rows for various met/chemistry
fields
> for
> > > different stats, and from there tabulate the day1, day2, and
day3
> > > performance for key regions. Also, the scorecards that I've
seen have
> > > red/green indicators with size of the symbols indicating a
greater
> > > difference between the models examined. Gray areas are not
reported.
> Is
> > > there a placeholder value that is assigned to areas in the table
that
> do
> > > not report any statistics? thanks.
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: building a scorecard using metviewer gui
From: Tatiana Burek
Time: Tue Mar 09 07:33:00 2021
Missing data would be displayed as an empty cell with the blue
background. In the default XML it is represented by:
<legend_range>
<color>#000000</color>
<background>#58ACFA</background>
<include_lower_limit>FALSE</include_lower_limit>
<include_upper_limit>FALSE</include_upper_limit>
<format_string>Not statistically relevant</format_string>
</legend_range>
Tatiana
On Mon Mar 08 15:44:47 2021, johnhg wrote:
> Ed (and Tatiana),
>
> I was hoping that the existing documentation would have had an
> explanation
> of that. But reading through this:
>
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
>
> I see a reference to this:
>
> *<threshold_file>:* full path to the XML file describing
> configurations for
> thresholds, colors and symbols (optional).
>
> But there's no further explanation of that threshold_file.
Presumably
> it'd
> contain options to specify how missing data cells are colored in. So
> we
> need more explanation and examples of that. I've cc'ed Tatiana on
this
> to
> let her know that we're missing documentation/examples of the
> threshold_file. I went poking around the METviewer repo but didn't
> immediately find any good info about that.
>
> Thanks,
> John
>
>
>
>
>
> On Mon, Mar 8, 2021 at 12:52 PM Edward Strobach - NOAA Affiliate via
> RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957 >
> >
> > Hi John,
> >
> > I was having a difficult time finding the right thing to say. I
knew
> > about
> > the size difference and their meaning, but what if there is
nothing
> > to
> > report on in one of the squares? Does that just get 'grayed' out?
> > I'll
> > give trial and error a try. Logan did send an xml template, but I
> > wanted
> > to confirm my understanding first. Here's my understanding as far
as
> > I
> > know it. Note that in Logan's original template he had only CSI
and
> > RMSE
> > for visibility and CAPE I believe. I'm taking largely from that
but
> > expanding it to fit my needs. Here's a copy of what I had sent if
> > you have
> > time to confirm my logic:
> >
> > Hi Logan,
> >
> > I think I understand most of the scorecard layout in the xml file.
> >
> > The template you sent creates rows first followed by columns. You
> > have
> > listed CSI, which would span, on the far left column, all the rows
> > where
> > thresholds are being reported. The next column over would be the
> > fields
> > (second column), followed by thresholds in the third column. You
> > also have
> > a second stat variable after CSI called FBIAS, although no
variables
> > appear
> > to be assigned there. This is what I've done so far:
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > * <rows> <field name="stat"> <val name="CSI"
> > label="Critical
> > Success Index"/> <field name="fcst_var"> <val
> > name="PMAVE"
> > label="Daily Ave. PM2.5"/> <field name="obs_thresh">
> > <val name="<10" label="< 10 ug/m3"/> <val
> > name="<20"
> > label="< 20 ug/m3"/> <val name="<30" label="<
30
> > ug/m3"/> <val name="<40" label="< 40 ug/m3"/>
> > <val name="<50" label="< 50 ug/m3"/> <val
> > name="<60"
> > label="< 60 ug/m3"/> </field> </field>
> > <field
> > name="fcst_var"> <val name="PDMAX1" label="Daily Max.
> > PM2.5"/>
> > <field name="obs_thresh"> <val name="<10"
> > label="<
> > 10 ug/m3"/> <val name="<20" label="< 20 ug/m3"/>
> > <val name="<30" label="< 30 ug/m3"/> <val
> > name="<40"
> > label="< 40 ug/m3"/> <val name="<50" label="<
50
> > ug/m3"/> <val name="<60" label="< 60 ug/m3"/>
> > </field> </field> <field name="fcst_var">
<val
> > name="OZMAX8" label="8-hour Max. Ozone"/> <field
> > name="obs_thresh"> <val name="<50" label="< 50
> > ppb"/>
> > <val name="<65" label="< 65 ppb"/> <val
> > name="<80" label="< 80 ppb"/> <val name="<100"
> > label="< 100 ppb"/> <val name="<125" label="<
125
> > ppb"/> <val name="<150" label="< 150 ppb"/>
> > </field> </field> <field name="fcst_var">
<val
> > name="OZMAX1" label="Hourly Max. Ozone"/> <field
> > name="obs_thresh"> <val name="<50" label="< 50
> > ppb"/>
> > <val name="<65" label="< 65 ppb"/> <val
> > name="<80" label="< 80 ppb"/> <val name="<100"
> > label="< 100 ppb"/> <val name="<125" label="<
125
> > ppb"/> <val name="<150" label="< 150 ppb"/>
> > </field>*
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > * </field> <val name="FBIAS" label="FBIAS"/>
> > <field
> > name="fcst_var"> <val name="PMTF" label="PM 2.5"/>
> > </field> <field name="fcst_var"> <val name="OZCON"
> > label="Ozone"/> </field> <field name="fcst_var">
> > <val name="TMP" label="2-m Temp"/> </field> <field
> > name="fcst_var"> <val name="DPT" label="2-m Dewpoint"/>
> > </field> <field name="fcst_var"> <val
name="WSPD10"
> > label="10-m Wind Speed"/> </field> <val name="RMSE"
> > label="RMSE"/> <field name="fcst_var"> <val
> > name="PMTF"
> > label="PM 2.5"/> </field> <field name="fcst_var">
> > <val name="OZCON" label="Ozone"/> </field> <field
> > name="fcst_var"> <val name="TMP" label="2-m Temp"/>
> > </field> <field name="fcst_var"> <val name="DPT"
> > label="2-m
> > Dewpoint"/> </field> <field name="fcst_var">
> > <val
> > name="WSPD10" label="10-m Wind Speed"/> </field>
</field>
> > </rows>*
> >
> > This would translate to something like this I think, or something
> > similar
> > (using the defined "label" of course):
> >
> > ________________________
> > | | <=10
> > | | <=20
> > | | <=30
> > CSI | PMAVE | <=40
> > * | | <=50*
> >
> > * | | <=60*
> >
> > * |___________________*
> >
> > * |*
> > I save the max/ave. fields for CSI and focus on the other fields
for
> > FBIAS
> > and RMSE. I marked in red what I was surprised by since I thought
> > that I
> > might have to close the val name with all fields of interest
> > contained
> > inside the xml block. In other words, I expected something like
this
> > instead:
> >
> > *<val name="RMSE" label="RMSE">*
> >
> >
> > * <field name="fcst_var"> <val name="PMTF"
label="PM
> > 2.5"/> </field>*
> > *....*
> > *....*
> > *</val>*
> >
> > For columns I believe I can do something similar. I can first
start
> > by
> > region (so have NWC, SWC, SEC, etc.)
> > Under each of these larger columns would be day1, day2, and day3,
> > i.e.
> >
> >
> > | NWC |
> > | day 1 | day 2 | day3 |
> >
> > In the past I was able to use a value name for forecast hour. In
> > this case
> > it seems that I would want to group 0-23, 24-47, and 48-71. I
think
> > this
> > grouping should be possible but I'm not sure about the appropriate
> > syntax
> > to do so. Also, if there's nothing to report, do I just Null it
out
> > in the
> > xml file or does that happen automatically.
> >
> > Hopefully I was clear in my email, but if I wasn't, then just let
me
> > know.
> > Thanks.
> >
> > On Mon, Mar 8, 2021 at 11:39 AM John Halley Gotway via RT <
> > met_help at ucar.edu>
> > wrote:
> >
> > > Ed,
> > >
> > > Unfortunately, no, there is no GUI interface for creating a
> > > scorecard. It
> > > must all be done with trial-and-error via the batch engine. If
you
> > > think
> > > this would be a very useful enhancement, you could recommend it
to
> > > Jason
> > > for inclusion in a future development and funding cycle.
> > >
> > > I'm not exactly sure what you mean by a placeholder value.
> > > Generally, the
> > > up/down pointing red/green arrows indicate statistically
> > > significant
> > > differences. In some cells, the differences may not be
> > > statistically
> > > significant. And that's different from having no data for a
cell.
> > > Here's
> > > the current documentation about scorecards:
> > >
> > >
https://dtcenter.github.io/METviewer/latest/Users_Guide/scorecard.html
> > >
> > > If you have more detailed questions about it, I could refer you
to
> > Tatiana.
> > >
> > > Hope that helps.
> > >
> > > Thanks,
> > > John
> > >
> > > On Mon, Mar 8, 2021 at 4:48 AM Edward Strobach - NOAA Affiliate
via
> > > RT <
> > > met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Mon Mar 08 04:48:26 2021: Request 98957 was acted upon.
> > > > Transaction: Ticket created by edward.strobach at noaa.gov
> > > > Queue: met_help
> > > > Subject: building a scorecard using metviewer gui
> > > > Owner: Nobody
> > > > Requestors: edward.strobach at noaa.gov
> > > > Status: new
> > > > Ticket <URL:
> > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=98957
> > >
> > > >
> > > >
> > > > Good morning,
> > > >
> > > > I have a scorecard xml template but would like to first build
a
> > scorecard
> > > > using the interactive metviewer gui so that I can optimize
what
> > > > I'd
> > like
> > > to
> > > > put in the scorecard. Can that be done? It's not clear that
it
> > > > can on
> > > the
> > > > gui. Ideally I'd like to build rows for various met/chemistry
> > > > fields
> > for
> > > > different stats, and from there tabulate the day1, day2, and
day3
> > > > performance for key regions. Also, the scorecards that I've
seen
> > > > have
> > > > red/green indicators with size of the symbols indicating a
> > > > greater
> > > > difference between the models examined. Gray areas are not
> > > > reported.
> > Is
> > > > there a placeholder value that is assigned to areas in the
table
> > > > that
> > do
> > > > not report any statistics? thanks.
> > > >
> > > >
> > >
> > >
> >
> >
------------------------------------------------
More information about the Met_help
mailing list