[Met_help] [rt.rap.ucar.edu #85847] History for Plotting RMSE time series?
Tatiana Burek via RT
met_help at ucar.edu
Tue Jun 26 13:44:03 MDT 2018
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hello. I'm trying to plot a time series of RMSEs - can you from the
attached xml if my set up is wrong? (no RMSE data in "R data").
Thanks -
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: Plotting RMSE time series?
From: Tatiana Burek
Time: Mon Jun 25 08:36:22 2018
Ying,
There are few problems with your xml.
- You specify two databases in the <database> tag -
mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1. mv_MODE_Jets_DJF2016_2017
database doesn't have any stat data and should not be used in this
case
- Your XML creates a graph using RMSE statistics stored in the
database: Statistics->Summary->None.
line_data_cnt table in mv_ylin_met_test1 database where such data is
stored is empty so you can't use this method. Instead you should chose
Statistics->Summary->SL1L2 or Statistics->Aggregation->SL1L2
- mv_ylin_met_test1 database contains data for mrms model only. There
is no data doe models cmorph and st4.
I created a valid plot of RMSE using your database and selecting
Statistics->Aggregation->SL1L2 : 20180625_142239
Tatiana
On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
> Hello. I'm trying to plot a time series of RMSEs - can you from the
> attached xml if my set up is wrong? (no RMSE data in "R data").
>
> Thanks -
>
> Ying
>
------------------------------------------------
Subject: Plotting RMSE time series?
From: Ying Lin
Time: Mon Jun 25 10:49:50 2018
Thank you very much Tatiana.
It wasn't the first time since MV started allowing multiple
databases that I had forgot to un-check the pre-existing one before
loading my own...
Thanks for the example. Oddly while plot_20180625_142239.png
appear to be the correct plot, plot_20180625_142239.xml seems to be
for
some other plot entirely. Nevertheless based on your email
descriptions (thanks!) I was able to re-create the plot.
MRMS has 'APCP' while CMORPH and ST4 have 'APCP_24' and I was
able
to make a separate RMSE plot for CMORPH and ST4 as well
(plot_20180625_161142.png).
Given that they have different variable names (APCP vs. APCP_24'
for the same field (24h accum of precip), is there a way to show all
three in one plot? Is Y1/Y2 the solution? Attached xml is my failed
attempt - would you be able to tell if there is a way to make this
work?
I've been trying to get 'MRMS' to produce a *.stat file with
'APCP_24' like the other two, so far no luck yet. Will submit a
separate ticket for that.
Thanks again,
Ying
On 06/25/2018 10:36 AM, Tatiana Burek via RT wrote:
> Ying,
>
> There are few problems with your xml.
> - You specify two databases in the <database> tag -
mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1. mv_MODE_Jets_DJF2016_2017
database doesn't have any stat data and should not be used in this
case
>
> - Your XML creates a graph using RMSE statistics stored in the
database: Statistics->Summary->None.
> line_data_cnt table in mv_ylin_met_test1 database where such data is
stored is empty so you can't use this method. Instead you should chose
Statistics->Summary->SL1L2 or Statistics->Aggregation->SL1L2
>
> - mv_ylin_met_test1 database contains data for mrms model only.
There is no data doe models cmorph and st4.
>
> I created a valid plot of RMSE using your database and selecting
Statistics->Aggregation->SL1L2 : 20180625_142239
>
> Tatiana
>
> On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
>> Hello. I'm trying to plot a time series of RMSEs - can you from
the
>> attached xml if my set up is wrong? (no RMSE data in "R data").
>>
>> Thanks -
>>
>> Ying
>>
>
>
--
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
Subject: Plotting RMSE time series?
From: Tatiana Burek
Time: Mon Jun 25 12:12:26 2018
Ying,
Take a look at 20180625_180848. I added both APCP_24 and APCP
variables and selected all models.
To reduce the clatter on the plot I hide the empty series.
Is it what you are looking for?
Tatiana
On Mon Jun 25 10:49:50 2018, ying.lin at noaa.gov wrote:
> Thank you very much Tatiana.
>
> It wasn't the first time since MV started allowing multiple
> databases that I had forgot to un-check the pre-existing one before
> loading my own...
>
> Thanks for the example. Oddly while plot_20180625_142239.png
> appear to be the correct plot, plot_20180625_142239.xml seems to be
> for
> some other plot entirely. Nevertheless based on your email
> descriptions (thanks!) I was able to re-create the plot.
>
> MRMS has 'APCP' while CMORPH and ST4 have 'APCP_24' and I was
> able
> to make a separate RMSE plot for CMORPH and ST4 as well
> (plot_20180625_161142.png).
>
> Given that they have different variable names (APCP vs. APCP_24'
> for the same field (24h accum of precip), is there a way to show all
> three in one plot? Is Y1/Y2 the solution? Attached xml is my
failed
> attempt - would you be able to tell if there is a way to make this
> work?
>
> I've been trying to get 'MRMS' to produce a *.stat file with
> 'APCP_24' like the other two, so far no luck yet. Will submit a
> separate ticket for that.
>
> Thanks again,
>
> Ying
>
> On 06/25/2018 10:36 AM, Tatiana Burek via RT wrote:
> > Ying,
> >
> > There are few problems with your xml.
> > - You specify two databases in the <database> tag -
> > mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1.
> > mv_MODE_Jets_DJF2016_2017 database doesn't have any stat data and
> > should not be used in this case
> >
> > - Your XML creates a graph using RMSE statistics stored in the
> > database: Statistics->Summary->None.
> > line_data_cnt table in mv_ylin_met_test1 database where such data
is
> > stored is empty so you can't use this method. Instead you should
> > chose Statistics->Summary->SL1L2 or Statistics->Aggregation->SL1L2
> >
> > - mv_ylin_met_test1 database contains data for mrms model only.
There
> > is no data doe models cmorph and st4.
> >
> > I created a valid plot of RMSE using your database and selecting
> > Statistics->Aggregation->SL1L2 : 20180625_142239
> >
> > Tatiana
> >
> > On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
> >> Hello. I'm trying to plot a time series of RMSEs - can you from
the
> >> attached xml if my set up is wrong? (no RMSE data in "R data").
> >>
> >> Thanks -
> >>
> >> Ying
> >>
> >
> >
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #85847] Plotting RMSE time series?
From: Ying Lin
Time: Mon Jun 25 12:42:46 2018
Hi Tatiana,
It's exactly what I'm looking for, thank you very much!
I loaded your xml file and it works ... but when you started on
this (w/o the existing xml file), how did you get all models to show
up? I selected APCP_24/RMSE and APCP/RMSE in Y1 dependent (forecast)
variables, and only cmorph and st4 showed up in the Y1 series
variables. When I reverse the order of Y1 dependent variables (APCP
on
top, followed by APCP_24), only mrms showed up in 'models'.
Thanks again -
Ying
On 06/25/2018 02:12 PM, Tatiana Burek via RT wrote:
> Ying,
>
> Take a look at 20180625_180848. I added both APCP_24 and APCP
variables and selected all models.
> To reduce the clatter on the plot I hide the empty series.
> Is it what you are looking for?
>
> Tatiana
> On Mon Jun 25 10:49:50 2018, ying.lin at noaa.gov wrote:
>> Thank you very much Tatiana.
>>
>> It wasn't the first time since MV started allowing multiple
>> databases that I had forgot to un-check the pre-existing one before
>> loading my own...
>>
>> Thanks for the example. Oddly while plot_20180625_142239.png
>> appear to be the correct plot, plot_20180625_142239.xml seems to be
>> for
>> some other plot entirely. Nevertheless based on your email
>> descriptions (thanks!) I was able to re-create the plot.
>>
>> MRMS has 'APCP' while CMORPH and ST4 have 'APCP_24' and I was
>> able
>> to make a separate RMSE plot for CMORPH and ST4 as well
>> (plot_20180625_161142.png).
>>
>> Given that they have different variable names (APCP vs.
APCP_24'
>> for the same field (24h accum of precip), is there a way to show
all
>> three in one plot? Is Y1/Y2 the solution? Attached xml is my
failed
>> attempt - would you be able to tell if there is a way to make this
>> work?
>>
>> I've been trying to get 'MRMS' to produce a *.stat file with
>> 'APCP_24' like the other two, so far no luck yet. Will submit a
>> separate ticket for that.
>>
>> Thanks again,
>>
>> Ying
>>
>> On 06/25/2018 10:36 AM, Tatiana Burek via RT wrote:
>>> Ying,
>>>
>>> There are few problems with your xml.
>>> - You specify two databases in the <database> tag -
>>> mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1.
>>> mv_MODE_Jets_DJF2016_2017 database doesn't have any stat data and
>>> should not be used in this case
>>>
>>> - Your XML creates a graph using RMSE statistics stored in the
>>> database: Statistics->Summary->None.
>>> line_data_cnt table in mv_ylin_met_test1 database where such data
is
>>> stored is empty so you can't use this method. Instead you should
>>> chose Statistics->Summary->SL1L2 or Statistics->Aggregation->SL1L2
>>>
>>> - mv_ylin_met_test1 database contains data for mrms model only.
There
>>> is no data doe models cmorph and st4.
>>>
>>> I created a valid plot of RMSE using your database and selecting
>>> Statistics->Aggregation->SL1L2 : 20180625_142239
>>>
>>> Tatiana
>>>
>>> On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
>>>> Hello. I'm trying to plot a time series of RMSEs - can you from
the
>>>> attached xml if my set up is wrong? (no RMSE data in "R data").
>>>>
>>>> Thanks -
>>>>
>>>> Ying
>>>>
>>>
>
>
--
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
Subject: Plotting RMSE time series?
From: Tatiana Burek
Time: Tue Jun 26 08:05:58 2018
Hi Ying,
In order to select all models for these two variables:
- open the web page and select mv_vlin_met_test1 database
- APCP is already selected -> choose the statistic (RMSE) and model
(only mrms is available)
- add a new variable and select APCP_24 and the statistic (RMSE). At
this point if you open a dropdown menu for the model values you still
would see only one mrms model.
- change s series variable from model to something else, for example
DESC, and than change it back to model.
It will refresh model values and you should see all three models on
the list.
Tatiana
On Mon Jun 25 12:42:46 2018, ying.lin at noaa.gov wrote:
> Hi Tatiana,
>
> It's exactly what I'm looking for, thank you very much!
>
> I loaded your xml file and it works ... but when you started on
> this (w/o the existing xml file), how did you get all models to show
> up? I selected APCP_24/RMSE and APCP/RMSE in Y1 dependent
(forecast)
> variables, and only cmorph and st4 showed up in the Y1 series
> variables. When I reverse the order of Y1 dependent variables (APCP
> on
> top, followed by APCP_24), only mrms showed up in 'models'.
>
> Thanks again -
>
> Ying
>
> On 06/25/2018 02:12 PM, Tatiana Burek via RT wrote:
> > Ying,
> >
> > Take a look at 20180625_180848. I added both APCP_24 and APCP
> > variables and selected all models.
> > To reduce the clatter on the plot I hide the empty series.
> > Is it what you are looking for?
> >
> > Tatiana
> > On Mon Jun 25 10:49:50 2018, ying.lin at noaa.gov wrote:
> >> Thank you very much Tatiana.
> >>
> >> It wasn't the first time since MV started allowing multiple
> >> databases that I had forgot to un-check the pre-existing one
before
> >> loading my own...
> >>
> >> Thanks for the example. Oddly while plot_20180625_142239.png
> >> appear to be the correct plot, plot_20180625_142239.xml seems to
be
> >> for
> >> some other plot entirely. Nevertheless based on your email
> >> descriptions (thanks!) I was able to re-create the plot.
> >>
> >> MRMS has 'APCP' while CMORPH and ST4 have 'APCP_24' and I
was
> >> able
> >> to make a separate RMSE plot for CMORPH and ST4 as well
> >> (plot_20180625_161142.png).
> >>
> >> Given that they have different variable names (APCP vs.
APCP_24'
> >> for the same field (24h accum of precip), is there a way to show
all
> >> three in one plot? Is Y1/Y2 the solution? Attached xml is my
> >> failed
> >> attempt - would you be able to tell if there is a way to make
this
> >> work?
> >>
> >> I've been trying to get 'MRMS' to produce a *.stat file with
> >> 'APCP_24' like the other two, so far no luck yet. Will submit a
> >> separate ticket for that.
> >>
> >> Thanks again,
> >>
> >> Ying
> >>
> >> On 06/25/2018 10:36 AM, Tatiana Burek via RT wrote:
> >>> Ying,
> >>>
> >>> There are few problems with your xml.
> >>> - You specify two databases in the <database> tag -
> >>> mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1.
> >>> mv_MODE_Jets_DJF2016_2017 database doesn't have any stat data
and
> >>> should not be used in this case
> >>>
> >>> - Your XML creates a graph using RMSE statistics stored in the
> >>> database: Statistics->Summary->None.
> >>> line_data_cnt table in mv_ylin_met_test1 database where such
data
> >>> is
> >>> stored is empty so you can't use this method. Instead you should
> >>> chose Statistics->Summary->SL1L2 or Statistics->Aggregation-
>SL1L2
> >>>
> >>> - mv_ylin_met_test1 database contains data for mrms model only.
> >>> There
> >>> is no data doe models cmorph and st4.
> >>>
> >>> I created a valid plot of RMSE using your database and selecting
> >>> Statistics->Aggregation->SL1L2 : 20180625_142239
> >>>
> >>> Tatiana
> >>>
> >>> On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
> >>>> Hello. I'm trying to plot a time series of RMSEs - can you
from
> >>>> the
> >>>> attached xml if my set up is wrong? (no RMSE data in "R
data").
> >>>>
> >>>> Thanks -
> >>>>
> >>>> Ying
> >>>>
> >>>
> >
> >
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #85847] Plotting RMSE time series?
From: Ying Lin
Time: Tue Jun 26 13:43:17 2018
Thanks Tatiana, I was able to follow the steps you outlined and
reproduce a plot I made yesterday starting from your xml file. This
is
good to know.
That's all on this now. Please close the ticket at your convenience.
Thank you again for being so helpful!
Ying
On 06/26/2018 10:05 AM, Tatiana Burek via RT wrote:
> Hi Ying,
> In order to select all models for these two variables:
> - open the web page and select mv_vlin_met_test1 database
> - APCP is already selected -> choose the statistic (RMSE) and model
(only mrms is available)
> - add a new variable and select APCP_24 and the statistic (RMSE). At
this point if you open a dropdown menu for the model values you still
would see only one mrms model.
> - change s series variable from model to something else, for example
DESC, and than change it back to model.
> It will refresh model values and you should see all three models on
the list.
>
> Tatiana
>
> On Mon Jun 25 12:42:46 2018, ying.lin at noaa.gov wrote:
>> Hi Tatiana,
>>
>> It's exactly what I'm looking for, thank you very much!
>>
>> I loaded your xml file and it works ... but when you started
on
>> this (w/o the existing xml file), how did you get all models to
show
>> up? I selected APCP_24/RMSE and APCP/RMSE in Y1 dependent
(forecast)
>> variables, and only cmorph and st4 showed up in the Y1 series
>> variables. When I reverse the order of Y1 dependent variables
(APCP
>> on
>> top, followed by APCP_24), only mrms showed up in 'models'.
>>
>> Thanks again -
>>
>> Ying
>>
>> On 06/25/2018 02:12 PM, Tatiana Burek via RT wrote:
>>> Ying,
>>>
>>> Take a look at 20180625_180848. I added both APCP_24 and APCP
>>> variables and selected all models.
>>> To reduce the clatter on the plot I hide the empty series.
>>> Is it what you are looking for?
>>>
>>> Tatiana
>>> On Mon Jun 25 10:49:50 2018, ying.lin at noaa.gov wrote:
>>>> Thank you very much Tatiana.
>>>>
>>>> It wasn't the first time since MV started allowing multiple
>>>> databases that I had forgot to un-check the pre-existing one
before
>>>> loading my own...
>>>>
>>>> Thanks for the example. Oddly while
plot_20180625_142239.png
>>>> appear to be the correct plot, plot_20180625_142239.xml seems to
be
>>>> for
>>>> some other plot entirely. Nevertheless based on your email
>>>> descriptions (thanks!) I was able to re-create the plot.
>>>>
>>>> MRMS has 'APCP' while CMORPH and ST4 have 'APCP_24' and I
was
>>>> able
>>>> to make a separate RMSE plot for CMORPH and ST4 as well
>>>> (plot_20180625_161142.png).
>>>>
>>>> Given that they have different variable names (APCP vs.
APCP_24'
>>>> for the same field (24h accum of precip), is there a way to show
all
>>>> three in one plot? Is Y1/Y2 the solution? Attached xml is my
>>>> failed
>>>> attempt - would you be able to tell if there is a way to make
this
>>>> work?
>>>>
>>>> I've been trying to get 'MRMS' to produce a *.stat file with
>>>> 'APCP_24' like the other two, so far no luck yet. Will submit a
>>>> separate ticket for that.
>>>>
>>>> Thanks again,
>>>>
>>>> Ying
>>>>
>>>> On 06/25/2018 10:36 AM, Tatiana Burek via RT wrote:
>>>>> Ying,
>>>>>
>>>>> There are few problems with your xml.
>>>>> - You specify two databases in the <database> tag -
>>>>> mv_MODE_Jets_DJF2016_2017,mv_ylin_met_test1.
>>>>> mv_MODE_Jets_DJF2016_2017 database doesn't have any stat data
and
>>>>> should not be used in this case
>>>>>
>>>>> - Your XML creates a graph using RMSE statistics stored in the
>>>>> database: Statistics->Summary->None.
>>>>> line_data_cnt table in mv_ylin_met_test1 database where such
data
>>>>> is
>>>>> stored is empty so you can't use this method. Instead you should
>>>>> chose Statistics->Summary->SL1L2 or Statistics->Aggregation-
>SL1L2
>>>>>
>>>>> - mv_ylin_met_test1 database contains data for mrms model only.
>>>>> There
>>>>> is no data doe models cmorph and st4.
>>>>>
>>>>> I created a valid plot of RMSE using your database and selecting
>>>>> Statistics->Aggregation->SL1L2 : 20180625_142239
>>>>>
>>>>> Tatiana
>>>>>
>>>>> On Sun Jun 24 17:30:00 2018, ying.lin at noaa.gov wrote:
>>>>>> Hello. I'm trying to plot a time series of RMSEs - can you
from
>>>>>> the
>>>>>> attached xml if my set up is wrong? (no RMSE data in "R
data").
>>>>>>
>>>>>> Thanks -
>>>>>>
>>>>>> 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