Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.graphics.apps.gnuplot > #4060 > unrolled thread

Best fit point in contour plots using Gnuplot

Started bysrishtinagu19@gmail.com
First post2018-11-19 00:20 -0800
Last post2018-11-20 23:10 -0800
Articles 13 — 2 participants

Back to article view | Back to comp.graphics.apps.gnuplot


Contents

  Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-19 00:20 -0800
    Re: Best fit point in contour plots using Gnuplot Karl Ratzsch <mail.kfr@gmx.net> - 2018-11-19 14:51 +0100
      Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-19 06:28 -0800
      Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-19 06:29 -0800
        Re: Best fit point in contour plots using Gnuplot Karl Ratzsch <mail.kfr@gmx.net> - 2018-11-19 16:27 +0100
        Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-19 07:29 -0800
          Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-20 01:00 -0800
            Re: Best fit point in contour plots using Gnuplot Karl Ratzsch <mail.kfr@gmx.net> - 2018-11-20 11:20 +0100
              Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-20 02:34 -0800
                Re: Best fit point in contour plots using Gnuplot Karl Ratzsch <mail.kfr@gmx.net> - 2018-11-20 13:06 +0100
                  Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-20 22:52 -0800
                  Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-20 22:52 -0800
                  Re: Best fit point in contour plots using Gnuplot srishtinagu19@gmail.com - 2018-11-20 23:10 -0800

#4060 — Best fit point in contour plots using Gnuplot

Fromsrishtinagu19@gmail.com
Date2018-11-19 00:20 -0800
SubjectBest fit point in contour plots using Gnuplot
Message-ID<f8b39595-16a3-4a02-b2a8-04c4d561fc60@googlegroups.com>
Hello Everyone,

Can someone please tell me how to make a best fit point in a contour plot
I am trying-
 $splot "test1.dat" with lines "<echo '135 0.40'"

But it is showing an error.
I want a point at 135,0.40


Thank you

[toc] | [next] | [standalone]


#4061

FromKarl Ratzsch <mail.kfr@gmx.net>
Date2018-11-19 14:51 +0100
Message-ID<psuf58$9lu$1@solani.org>
In reply to#4060
Am 19.11.2018 um 09:20 schrieb srishtinagu19@gmail.com:

> Can someone please tell me how to make a best fit point in a contour plot
> I am trying-
>  $splot "test1.dat" with lines "<echo '135 0.40'"

What gave you the idea that this could work? ;-)

> But it is showing an error.
> I want a point at 135,0.40

Two possiblities

   set object 1 ellipse at 134,0.4 size 2,2 units xx

or

   set label 1 "" at 135,0.4 point pt 3 ps 2

followed by your (s)plot command

[toc] | [prev] | [next] | [standalone]


#4062

Fromsrishtinagu19@gmail.com
Date2018-11-19 06:28 -0800
Message-ID<dcdcdc9f-bea9-4be3-b3d1-a836da323d93@googlegroups.com>
In reply to#4061
Thank you.. Its making now..
echo command i saw somewhere, where it was used to make a point so i tried to combine it with splot command
Thanks..

[toc] | [prev] | [next] | [standalone]


#4063

Fromsrishtinagu19@gmail.com
Date2018-11-19 06:29 -0800
Message-ID<7f45f7f2-7355-4a87-86c8-100de899e262@googlegroups.com>
In reply to#4061

There is one more issue in my contour plot...
The axes are not proper... 
Both the axes should have the same origin point... the y axis is on the vertically flipped side of x axis
How can I resolve it ...

[toc] | [prev] | [next] | [standalone]


#4064

FromKarl Ratzsch <mail.kfr@gmx.net>
Date2018-11-19 16:27 +0100
Message-ID<psukob$dek$1@solani.org>
In reply to#4063
Am 19.11.2018 um 15:29 schrieb srishtinagu19@gmail.com:

> There is one more issue in my contour plot...
> The axes are not proper... 
> Both the axes should have the same origin point... the y axis is on the vertically flipped side of x axis
> How can I resolve it ...

Please show your actual set of gnuplot commands. I'm neither clairvoyant
nor standing right behind you.



[toc] | [prev] | [next] | [standalone]


#4065

Fromsrishtinagu19@gmail.com
Date2018-11-19 07:29 -0800
Message-ID<e5bb71c8-8e5f-4eae-975d-12b68417961e@googlegroups.com>
In reply to#4063
gnuplot> set contour
gnuplot> set view 0,0
gnuplot> unset surface
gnuplot> set cntrparam levels discrete 2.3,6.2,11.8
gnuplot> splot[0:360][0:0.9] "test1.dat"

[toc] | [prev] | [next] | [standalone]


#4066

Fromsrishtinagu19@gmail.com
Date2018-11-20 01:00 -0800
Message-ID<d08c30c8-ef36-4e21-a14c-e335775cb058@googlegroups.com>
In reply to#4065
On Monday, November 19, 2018 at 8:59:43 PM UTC+5:30, srisht...@gmail.com wrote:
> gnuplot> set contour
> gnuplot> set view 0,0
> gnuplot> unset surface
> gnuplot> set cntrparam levels discrete 2.3,6.2,11.8
> gnuplot> splot[0:360][0:0.9] "test1.dat"

Please tell how can I make my axes properly.. I want to show my plot... how should i post it here ? I dont find any option of attaching a file  

[toc] | [prev] | [next] | [standalone]


#4067

FromKarl Ratzsch <mail.kfr@gmx.net>
Date2018-11-20 11:20 +0100
Message-ID<pt0n56$p5k$1@solani.org>
In reply to#4066
Am 20.11.2018 um 10:00 schrieb srishtinagu19@gmail.com:
> On Monday, November 19, 2018 at 8:59:43 PM UTC+5:30, srisht...@gmail.com wrote:
>> gnuplot> set contour
>> gnuplot> set view 0,0
>> gnuplot> unset surface
>> gnuplot> set cntrparam levels discrete 2.3,6.2,11.8
>> gnuplot> splot[0:360][0:0.9] "test1.dat"
> 
> Please tell how can I make my axes properly.. I want to show my plot... how should i post it here ? I dont find any option of attaching a file  
> 

I think what you want is "set view map", to get an actual 2D contour
plot instead of viewing the 3D plot in the z direction.


set contour
unset surface
set view map
splot ...

[toc] | [prev] | [next] | [standalone]


#4068

Fromsrishtinagu19@gmail.com
Date2018-11-20 02:34 -0800
Message-ID<39d92ec1-e638-4fa2-b513-7d1afceb5b19@googlegroups.com>
In reply to#4067
On Tuesday, November 20, 2018 at 3:50:23 PM UTC+5:30, Karl Ratzsch wrote:
> Am 20.11.2018 um 10:00 schrieb srishtinagu19@gmail.com:
> > On Monday, November 19, 2018 at 8:59:43 PM UTC+5:30, srisht...@gmail.com wrote:
> >> gnuplot> set contour
> >> gnuplot> set view 0,0
> >> gnuplot> unset surface
> >> gnuplot> set cntrparam levels discrete 2.3,6.2,11.8
> >> gnuplot> splot[0:360][0:0.9] "test1.dat"
> > 
> > Please tell how can I make my axes properly.. I want to show my plot... how should i post it here ? I dont find any option of attaching a file  
> > 
> 
> I think what you want is "set view map", to get an actual 2D contour
> plot instead of viewing the 3D plot in the z direction.
> 
> 
> set contour
> unset surface
> set view map
> splot ...

.... I just tried it set view map... it worked... Thank you.

Also I want to know does gnuplot automatically calculates the 1,2,3.. sigma ranges ?
I have three columns.. x,y,z. For contour plot I am using the above commands with -
 "$set cntrparam levels discrete 2.3,6.2,11.8"
 which are 1,2 and 3 sigma levels respectively for 2 parameter correlation.
But in my z column these sigma values are not present still I am getting a plot.. I am confused that what does it mean ? 
What I know is- for particular z values of confidence levels contours are drawn for x and y for corresponding z values set by the user.. Am I correct ?

[toc] | [prev] | [next] | [standalone]


#4070

FromKarl Ratzsch <mail.kfr@gmx.net>
Date2018-11-20 13:06 +0100
Message-ID<pt0tcs$t8b$1@solani.org>
In reply to#4068
Am 20.11.2018 um 11:34 schrieb srishtinagu19@gmail.com:
> ..... I just tried it set view map... it worked... Thank you.
> 
> Also I want to know does gnuplot automatically calculates the 1,2,3.. sigma ranges ?
> I have three columns.. x,y,z. For contour plot I am using the above commands with -
>  "$set cntrparam levels discrete 2.3,6.2,11.8"
>  which are 1,2 and 3 sigma levels respectively for 2 parameter correlation.

If you say so ...

> But in my z column these sigma values are not present still I am getting a plot.. I am confused that what does it mean ? 

gnuplot simply makes a surface with height z out of your xy-gridded
datapoints, and then interpolates contour lines onto that at the
specified levels. Neither gnuplot nor I know what "sigma" is.

> What I know is- for particular z values of confidence levels contours are drawn for x and y for corresponding z values set by the user.. Am I correct ?
> 

Confidence levels of what? I repeat, I am NOT clairvoyant. I cannot read
your mind. Contours are made by a straightforward interpolation, there
is no stochastic element.

[toc] | [prev] | [next] | [standalone]


#4079

Fromsrishtinagu19@gmail.com
Date2018-11-20 22:52 -0800
Message-ID<4d1411e3-1bec-4d72-ae32-71e967184400@googlegroups.com>
In reply to#4070
On Tuesday, November 20, 2018 at 5:36:54 PM UTC+5:30, Karl Ratzsch wrote:
> Am 20.11.2018 um 11:34 schrieb srishtinagu19@gmail.com:
> > ..... I just tried it set view map... it worked... Thank you.
> > 
> > Also I want to know does gnuplot automatically calculates the 1,2,3.. sigma ranges ?
> > I have three columns.. x,y,z. For contour plot I am using the above commands with -
> >  "$set cntrparam levels discrete 2.3,6.2,11.8"
> >  which are 1,2 and 3 sigma levels respectively for 2 parameter correlation.
> 
> If you say so ...
> 
> > But in my z column these sigma values are not present still I am getting a plot.. I am confused that what does it mean ? 
> 
> gnuplot simply makes a surface with height z out of your xy-gridded
> datapoints, and then interpolates contour lines onto that at the
> specified levels. Neither gnuplot nor I know what "sigma" is.
> 
> > What I know is- for particular z values of confidence levels contours are drawn for x and y for corresponding z values set by the user.. Am I correct ?
> > 
> 
> Confidence levels of what? I repeat, I am NOT clairvoyant. I cannot read
> your mind. Contours are made by a straightforward interpolation, there
> is no stochastic element.



>set contour
>set view map
>unset surface
>set cntrparam levels discrete 2.3,6.2,11.8
>splot [0:360][0:0.9] "test1.dat" with lines


These are the commands that I am using.
I simply want to know how to plot contours at various confidence levels.
By Sigma, I mean - confidence level of a data like 1 sigma means result which is accurate within 68% of its true value 2.3 and so on...

[toc] | [prev] | [next] | [standalone]


#4080

Fromsrishtinagu19@gmail.com
Date2018-11-20 22:52 -0800
Message-ID<9405a864-6815-4b34-8ee8-f47bc866066f@googlegroups.com>
In reply to#4070
On Tuesday, November 20, 2018 at 5:36:54 PM UTC+5:30, Karl Ratzsch wrote:
> Am 20.11.2018 um 11:34 schrieb srishtinagu19@gmail.com:
> > ..... I just tried it set view map... it worked... Thank you.
> > 
> > Also I want to know does gnuplot automatically calculates the 1,2,3.. sigma ranges ?
> > I have three columns.. x,y,z. For contour plot I am using the above commands with -
> >  "$set cntrparam levels discrete 2.3,6.2,11.8"
> >  which are 1,2 and 3 sigma levels respectively for 2 parameter correlation.
> 
> If you say so ...
> 
> > But in my z column these sigma values are not present still I am getting a plot.. I am confused that what does it mean ? 
> 
> gnuplot simply makes a surface with height z out of your xy-gridded
> datapoints, and then interpolates contour lines onto that at the
> specified levels. Neither gnuplot nor I know what "sigma" is.
> 
> > What I know is- for particular z values of confidence levels contours are drawn for x and y for corresponding z values set by the user.. Am I correct ?
> > 
> 
> Confidence levels of what? I repeat, I am NOT clairvoyant. I cannot read
> your mind. Contours are made by a straightforward interpolation, there
> is no stochastic element.



>set contour
>set view map
>unset surface
>set cntrparam levels discrete 2.3,6.2,11.8
>splot [0:360][0:0.9] "test1.dat" with lines


These are the commands that I am using.
I simply want to know how to plot contours at various confidence levels.
By Sigma, I mean - confidence level of a data like 1 sigma means result which is accurate within 68% of its true value 2.3 and so on...

[toc] | [prev] | [next] | [standalone]


#4081

Fromsrishtinagu19@gmail.com
Date2018-11-20 23:10 -0800
Message-ID<1a791c87-48a7-49ed-8342-85a4065120cf@googlegroups.com>
In reply to#4070
On Tuesday, November 20, 2018 at 5:36:54 PM UTC+5:30, Karl Ratzsch wrote:
> Am 20.11.2018 um 11:34 schrieb srishtinagu19@gmail.com:
> > ..... I just tried it set view map... it worked... Thank you.
> > 
> > Also I want to know does gnuplot automatically calculates the 1,2,3.. sigma ranges ?
> > I have three columns.. x,y,z. For contour plot I am using the above commands with -
> >  "$set cntrparam levels discrete 2.3,6.2,11.8"
> >  which are 1,2 and 3 sigma levels respectively for 2 parameter correlation.
> 
> If you say so ...
> 
> > But in my z column these sigma values are not present still I am getting a plot.. I am confused that what does it mean ? 
> 
> gnuplot simply makes a surface with height z out of your xy-gridded
> datapoints, and then interpolates contour lines onto that at the
> specified levels. Neither gnuplot nor I know what "sigma" is.
> 
> > What I know is- for particular z values of confidence levels contours are drawn for x and y for corresponding z values set by the user.. Am I correct ?
> > 
> 
> Confidence levels of what? I repeat, I am NOT clairvoyant. I cannot read
> your mind. Contours are made by a straightforward interpolation, there
> is no stochastic element.


I have understood your explanation...
Thank you.

[toc] | [prev] | [standalone]


Back to top | Article view | comp.graphics.apps.gnuplot


csiph-web