gnuplot-info — For questions and discussion about gnuplot

[Gnuplot-info] quality of fill pattern from one term to another

[Mahmood N. <nt_...@ya...>]

Hello,
It seems that the fill patterns quality are changed from one terminal to another. For example the 


   fill pattern 4 lc rgb "black"

on the windows terminal is a thick '\\' shape. However when I set the terminal to 


   set term post eps enhanced color blacktext size 3.2,2.2 solid "Times-Roman" 12

The pattern is thin '\\' shape.

Is there any workaround for that?

 
Regards,
Mahmood

Re: [Gnuplot-info] Proper gaussian weight for smooth kdensity with logscale x?

[Rhys U. <rhy...@gm...>]

> When plotting with 'smooth kdensity' I must provide 1/number-of-points
> as the second plotting column to get a normalized histogram.  Does
> that scaling factor need to change to get a properly normalized
> histogram when using a logarithmic x axis?

While not The Right Thing (TM), here's what I did in case someone else
finds it useful:

# First, use 'set table' to get the proper kdens curve fit
set samples 1000
set table 'random.kdens'
plot "random.tmp" using 1:(1.0/nsamp) title 'Logscale x' smooth kdensity
unset table

# Second, plot the curve fit on a logscale
set output "logscale.ps"
set logscale x
plot "random.kdens" title 'Logscale x' smooth kdensity with lines

Please give a shout if there's something cleaner/prettier/simpler,
Rhys

[Gnuplot-info] How to get graph coordinate?

[smms <sm...@ho...>]

I have seven plots each with different x and y ranges. I need to put a
vertical line in each plot at a fixed graph coordinate. I understand to do
that I need to know the graph coordinates. For example, for one plot, I have
the following ranges and I am trying to add a vertical line by,

set xrange [0:10]
set yrange [0:100]
set arrow from first 2, graph 0 to first 2, graph 1 nohead

But for a different plot I cant use the same set arrow option due to varying
first <value>. How can I get the equivalent graph coordinate to solve this
problem? 

Any help will be appreciated.





--
View this message in context: http://gnuplot.10905.n7.nabble.com/How-to-get-graph-coordinate-tp17905.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Any suggestion for better fit function?

[smms <sm...@ho...>]

Many thanks Dietar for sparing your time for me. 
Eventually I managed to fit with g(x) = a*atan(x/b).



--
View this message in context: http://gnuplot.10905.n7.nabble.com/Re-Any-suggestion-for-better-fit-function-tp17900p17904.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Proper gaussian weight for smooth kdensity with logscale x?

[Rhys U. <rhy...@gm...>]

When plotting with 'smooth kdensity' I must provide 1/number-of-points
as the second plotting column to get a normalized histogram.  Does
that scaling factor need to change to get a properly normalized
histogram when using a logarithmic x axis?  I should be able to sort
through this myself, but my head's not working today.

More concretely, should the (1.0/nsamp) factor be modified in some
fashion to obtain a normalized kernel density estimate for the second
plot below?

set print "random.tmp"
mean  = 5
nsamp = 10000
do for [i=1:nsamp] {
    print sprintf("%8.5g", invnorm(rand(0)) + mean)
}
unset print

set term postscript

# Normalized per 'help smooth kdensity'
set output "regular.ps"
plot "random.tmp" using 1:(1.0/nsamp) smooth kdensity

# How should this be normalized?
set logscale x
set output "logscale.ps"
plot "random.tmp" using 1:(1.0/nsamp) title 'Logscale x' smooth kdensity

Thank you,
Rhys

Re: [Gnuplot-info] v4.6 forces X scale text off the bottom of the canvas

[Matthew J. <nag...@ed...>]

I have found a workaround that work for me.

Add
set xlabel " "

I found this at
https://github.com/OpenTSDB/opentsdb/issues/107


On 6/01/2014 9:51 PM, Matthew Jurgens wrote:
> I was previously running v4.4 and all was working perfectly. I upgraded
> to v4.6.1 on Fedora 19 and now the X scale text is truncated off the
> bottom of the canvas.
> See and example at: http://www.edcint.co.nz/tmp/test.png
>
> Any ideas what is causing this now or how to fix it?
>
> I'm graphing with the following:
>
> set grid noxtics ytics
> set nokey
> set size 1,1
> set terminal png transparent font DejaVuSansMono 8 size 448,196
> set xtics rotate
> set xdata time
> set timefmt "%Y%m%d:%H%M%S"
> set format x " %R"
> set y2tics
> set format y2
> set format y
>
> set xrange ["20140105:212400":"20140106:205400"]
>
> # 7 day styles and older, temp, humid, pressure, rain
> set style line 1 linewidth 1 linecolor rgbcolor "cyan"
> set style line 2 linewidth 1 linecolor rgbcolor "brown"
> set style line 3 linewidth 1 linecolor rgbcolor "green"
> set style line 4 linewidth 1 linecolor rgbcolor "orange"
> # max temp
> set style line 9 linewidth 1 linecolor rgbcolor "red"
>
> # 1 day styles , temp, humid, pressure, rain
> set style line 5 linewidth 1 linecolor rgbcolor "blue"
> set style line 6 linewidth 1 linecolor rgbcolor "orange"
> set style line 7 linewidth 1 linecolor rgbcolor "sea-green"
> set style line 8 linewidth 1 linecolor rgbcolor "plum"
> # max temp
> set style line 10 linewidth 1 linecolor rgbcolor "magenta"
>
> plot "/tmp/test/Adelaide.data" using 1:2:(0.01) smooth acsplines ls 5
> ,"/tmp/test/Adelaide.data" using 1:6 with lines ls 10
>
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT
> organizations don't have a clear picture of how application performance
> affects their revenue. With AppDynamics, you get 100% visibility into your
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>


-- 
Smartmon System Monitoring <http://www.smartmon.com.au>
www.smartmon.com.au <http://www.smartmon.com.au>

[Gnuplot-info] v4.6 forces X scale text off the bottom of the canvas

[Matthew J. <nag...@ed...>]

I was previously running v4.4 and all was working perfectly. I upgraded 
to v4.6.1 on Fedora 19 and now the X scale text is truncated off the 
bottom of the canvas.
See and example at: http://www.edcint.co.nz/tmp/test.png

Any ideas what is causing this now or how to fix it?

I'm graphing with the following:

set grid noxtics ytics
set nokey
set size 1,1
set terminal png transparent font DejaVuSansMono 8 size 448,196
set xtics rotate
set xdata time
set timefmt "%Y%m%d:%H%M%S"
set format x " %R"
set y2tics
set format y2
set format y

set xrange ["20140105:212400":"20140106:205400"]

# 7 day styles and older, temp, humid, pressure, rain
set style line 1 linewidth 1 linecolor rgbcolor "cyan"
set style line 2 linewidth 1 linecolor rgbcolor "brown"
set style line 3 linewidth 1 linecolor rgbcolor "green"
set style line 4 linewidth 1 linecolor rgbcolor "orange"
# max temp
set style line 9 linewidth 1 linecolor rgbcolor "red"

# 1 day styles , temp, humid, pressure, rain
set style line 5 linewidth 1 linecolor rgbcolor "blue"
set style line 6 linewidth 1 linecolor rgbcolor "orange"
set style line 7 linewidth 1 linecolor rgbcolor "sea-green"
set style line 8 linewidth 1 linecolor rgbcolor "plum"
# max temp
set style line 10 linewidth 1 linecolor rgbcolor "magenta"

plot "/tmp/test/Adelaide.data" using 1:2:(0.01) smooth acsplines ls 5 
,"/tmp/test/Adelaide.data" using 1:6 with lines ls 10

Re: [Gnuplot-info] Any suggestion for better fit function?

[Dr. I. D. J. <die...@t-...>]

Well,
kind' of "did-it-for-you" ... :-)

What about g(x)=a*x/(1.0+b*x)*exp(c*x)

through

fit g(x) "test.dat" via a,b,c

results here in 
a=68.4844227089402
b=23.4451718533986
c=-0.0329804676514508

and the fit seems quite acceptable (for me ...)

HTH,
take care




Dieter Jurzitza

On Donnerstag, 2. Januar 2014 23:16:05 smms wrote:
> I know it is rather a "do-it-for-me" post but thought someone could help me
************
-- 
-----------------------------------------------------------

                               |
                                \
                 /\_/\           |
                | ~x~ |/-----\   /
                 \   /-       \_/
  ^^__   _        /  _  ____   /
 <°°__ \- \_/     |  |/    |  |
  ||  ||         _| _|    _| _|

if you really want to see the pictures above - use some font
with constant spacing like courier! :-)
-----------------------------------------------------------

[Gnuplot-info] Any suggestion for better fit function?

[smms <sm...@ho...>]

I know it is rather a "do-it-for-me" post but thought someone could help me
with a better fit suggestion. 

Here is my data set I managed to fit (not quite exciting) with following
scripts and function. Just looking for if someone could suggest a better
function. Many thanks for any feedback.

reset
set key bottom right
f(x) = a*atan(x/b); a = 2.65; b = 2.5 
fit f(x) 'test.dat' u 4:25 via a,b
plot 'test.dat' u 4:25 w p not '0.99-3%-1200', f(x) t 'f(x)'


test.dat data set is here,

2.561	2.642
0.686	2.685
0.586	2.647
0.487	2.658
0.402	2.620
0.284	2.516
0.213	2.440
0.197	2.359

Here is my fitted plot,

<http://gnuplot.10905.n7.nabble.com/file/n17899/test.bmp> 




--
View this message in context: http://gnuplot.10905.n7.nabble.com/Any-suggestion-for-better-fit-function-tp17899.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Can I manually change the labels on the horizontal axis?

[smms <sm...@ho...>]

set xlabel 'text'
set ylabel 'text'



--
View this message in context: http://gnuplot.10905.n7.nabble.com/Can-I-manually-change-the-labels-on-the-horizontal-axis-tp17893p17898.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Writing additional text on axis

[smms <sm...@ho...>]

I guess you can take advantage of the following options:

set label "text" at x,y
set arrow from x,y to x1,y1 no head



--
View this message in context: http://gnuplot.10905.n7.nabble.com/Writing-additional-text-on-axis-tp17895p17897.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Writing additional text on axis

[Mahmood N. <nt_...@ya...>]

Hi
I am trying to put some additional information on the x-axis like this

----+----+----+----+----+----+----+----+
    1    2    3    4    5    6    7    8
<---Interval #1---><---Interval #2--->

Where should I look for such formatting?

 
Regards,
Mahmood

Re: [Gnuplot-info] Clustered histogram and the location of the first cluster?

[Ethan A M. <eam...@gm...>]

On Friday, 20 December 2013 01:39:17 PM Risto Toivonen wrote:
> Hi,
> 
> I came across a problem when using the histogram cluster style.
> Manual states that the location of the first cluster at the x-axis is
> x=1. However, if I use explicit title option then the first cluster is
> located at x=0. Is there way to get an equal behavior for the both of
> the situations (with and without the title)?

The documentation may be confusing.
Here is what is happening:

Your data file contains header information, not data values,
on the first line.  That's fine, but the program doesn't magically know that.
It needs some hint from user that the first line is special.

>  plot for[I=2:3] 'test.dat' using I:xticlabels(1) title columnheader(I)

Here "title columnheader()" clues in the program that the first line is
special.  The first line is used to generate labels and the following line
(2nd line of file) is used a the first row of data   y1[0] y2[0] ...
Because the index is 0, the first histogram cluster is at x=0

> plot for[I=2:3] 'test.dat' using I:xticlabels(1) notitle

Here there is no hint that the first line is special, so the program
tries to read it as data.  But the contents of the first line aren't
readable as numbers so y1[0] y2[0] ... produce no histogram
bars.  The first visible set of bars is y1[1] y2[1] which is drawn at x=1.
If you replaces labels "M1" and "M2" with "1" and "2" you'll see 
what's happening.

The solution to your problem is to give the program a global
hint that the first line of the input contains labels, even though
the plot command may not use them:

 set key autotitle columnheader

	Ethan



> 
> My test file (test.dat):
> -------------
> part 	M1 	M2	p
> A	2.42	2.2	0.01
> B	2.50	2.1	0.03
> -------------
> 
> Commands:
> -------------
> reset
> set style data histograms
> set style histogram cluster
> set style fill solid 1.0 border lt -1
> set key left top
> set xrange [-0.5:2.5]
> set yrange [0:3]
> 
> plot for[I=2:3] 'test.dat' using I:xticlabels(1) notitle
> # Example here: http://picpaste.com/test-notitle-1u9GkFC8.png
> 
> plot for[I=2:3] 'test.dat' using I:xticlabels(1) title columnheader(I)
> # Example here: http://picpaste.com/test-title-fAPyp44x.png
> -------------
> 
> I'm using the following: G N U P L O T Version 4.6 patchlevel 3 last
> modified 2013-04-12 Build System: FreeBSD i386
> 
> 
> Any ideas?
> 
> Thank you,
> Risto
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT 
> organizations don't have a clear picture of how application performance 
> affects their revenue. With AppDynamics, you get 100% visibility into your 
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info

Re: [Gnuplot-info] Install gnuplot 4.6.3 on Mac OS-X

[Tom L. <tl...@cp...>]

Personally, I use homebrew: http://brew.sh/

-It seems to be the least intrusive of the package mangers, and uses the current gnuplot. Once you’ve got homebrew, simply

brew install gnuplot

-though I did have to do something that I’ve now forgotten to make sure I got freetype, for scalable fonts… I posted the answer online somewhere, but forget where! Ahh- the wonders of bash history- here we go:

brew install libgd --with-freetype

Hope that helps!

~Tom


On 20 Dec 2013, at 10:35, fre...@li... wrote:

> Hello,I am having troubles in installing the gnuplot 4.6.3 for Mac OS-X 10.8.5.The step-by-step guides that I found online are a bit confusing.Is there a simple procedure to follow, once I download the package from http://sourceforge.net/projects/gnuplot/files/?Thanks
> Federico 
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT 
> organizations don't have a clear picture of how application performance 
> affects their revenue. With AppDynamics, you get 100% visibility into your 
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info

[Gnuplot-info] Install gnuplot 4.6.3 on Mac OS-X

[<fre...@li...>]

Hello,I am having troubles in installing the gnuplot 4.6.3 for Mac OS-X 10.8.5.The step-by-step guides that I found online are a bit confusing.Is there a simple procedure to follow, once I download the package from http://sourceforge.net/projects/gnuplot/files/?Thanks
Federico 

[Gnuplot-info] Clustered histogram and the location of the first cluster?

[Risto T. <toi...@ne...>]

Hi,

I came across a problem when using the histogram cluster style.
Manual states that the location of the first cluster at the x-axis is
x=1. However, if I use explicit title option then the first cluster is
located at x=0. Is there way to get an equal behavior for the both of
the situations (with and without the title)?

My test file (test.dat):
-------------
part 	M1 	M2	p
A	2.42	2.2	0.01
B	2.50	2.1	0.03
-------------

Commands:
-------------
reset
set style data histograms
set style histogram cluster
set style fill solid 1.0 border lt -1
set key left top
set xrange [-0.5:2.5]
set yrange [0:3]

plot for[I=2:3] 'test.dat' using I:xticlabels(1) notitle
# Example here: http://picpaste.com/test-notitle-1u9GkFC8.png

plot for[I=2:3] 'test.dat' using I:xticlabels(1) title columnheader(I)
# Example here: http://picpaste.com/test-title-fAPyp44x.png
-------------

I'm using the following: G N U P L O T Version 4.6 patchlevel 3 last
modified 2013-04-12 Build System: FreeBSD i386


Any ideas?

Thank you,
Risto

Re: [Gnuplot-info] Which Symbol font in Postscript terminal?

[Ethan M. <eam...@gm...>]

I think there is a misunderstanding inherent in your question.  When
gnuplot produces a PostScript file (including *.eps), by default it simply
writes "Symbol" as a requested font name.  The choice of what font is
actully used to satisfy this request is made by the program that renders
the file.  Gnuplot has no control over this.

You do have the option of embedding a font in the file produced by gnuplot.
  This is done using the "fontfile <fontfilename>" option to "set term
post".  However, you still have to make sure that you indeed provide a file
that is the true Adobe Symbol font using Adobe's special symbol encoding,
rather than some other font that just happens to be named "Symbol".



On Sat, Dec 7, 2013 at 7:35 AM, Nunzio Losacco <nun...@gm...>wrote:

> Hi all,
>
> I am wondering which Symbol font the Postscript terminal is using. I am
> producing .eps files and importing them in MS Word for Mac (the journal I'm
> submitting my paper to only accepts manuscripts in MS Word) and the Symbol
> characters sometimes disappear when the file is printed in PDF during the
> submission procedure. Thus I think the only option to get always a good
> result is to embed the fonts in the file through "fontfile add", which is
> what I currently do with other fonts. But how do I know which fonts in my
> system are called by the "Symbol" alias in  gnuplot?
> Can you help me with that?
>
> Regards,
>
> N.
>
> ------------------------------------------------------------------------------
> Sponsored by Intel(R) XDK
> Develop, test and display web and hybrid apps with a single code base.
> Download it for free now!
>
> http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk
> _______________________________________________
> gnuplot-info mailing list
> gnu...@li...
> Membership management via:
> https://lists.sourceforge.net/lists/listinfo/gnuplot-info
>

[Gnuplot-info] Problem: Bar plot with 3 groups

[<gp...@bl...>]

Hi All,
I am having problems getting what I want. I would like to have 3 (or more) groups of bars, each in their own colour. Each group shows the same points over time. I don't want to mix them in the normal way. Ideally I would like to have the points in each group numbered 1,2,etc and the groups with colours in a key. However, that didn't seem possible.
Here is my example data file:
Time1;15;11;12;13;11;10;10;11;15;11;12
Time2;8;9;8;7;8;10;11;8;9;12;13
Time3;4;5;6;4;3;4;5;6;7;2;3
When I use this code (I get more or less what I want but not in different colours):
set auto x
set yrange [0:16]
set ylabel 'Number'
set xlabel 'Group'
set style data histogram
set style histogram cluster gap 5
set style fill solid border rgb "black"
unset key
set datafile separator ";"
plot \
'data.dat' using 2:xtic(1) ls 1, \
'data.dat' using 3:xtic(1) ls 1, \
'data.dat' using 4:xtic(1) ls 1, \
'data.dat' using 5:xtic(1) ls 1, \
'data.dat' using 6:xtic(1) ls 1, \
'data.dat' using 7:xtic(1) ls 1, \
'data.dat' using 8:xtic(1) ls 1, \
'data.dat' using 9:xtic(1) ls 1, \
'data.dat' using 10:xtic(1) ls 1, \
'data.dat' using 11:xtic(1) ls 1, \
'data.dat' using 12:xtic(1) ls 1
So I tried this:
set auto x
set yrange [0:16]
set ylabel 'Number'
set xlabel 'Group'
set style data histogram
set style histogram cluster gap 5
set style fill solid border rgb "black"
unset key
set datafile separator ";"
set style line 1 lc rgb 'blue'  
set style line 2 lc rgb 'green' 
set style line 3 lc rgb 'yellow' 
plot \
'data.dat' every ::0::0 using 2:xtic(1) ls 1, \
'data.dat' every ::0::0 using 3:xtic(1) ls 1, \
'data.dat' every ::0::0 using 4:xtic(1) ls 1, \
'data.dat' every ::0::0 using 5:xtic(1) ls 1, \
'data.dat' every ::0::0 using 6:xtic(1) ls 1, \
'data.dat' every ::0::0 using 7:xtic(1) ls 1, \
'data.dat' every ::0::0 using 8:xtic(1) ls 1, \
'data.dat' every ::0::0 using 9:xtic(1) ls 1, \
'data.dat' every ::0::0 using 10:xtic(1) ls 1, \
'data.dat' every ::0::0 using 11:xtic(1) ls 1, \
'data.dat' every ::0::0 using 12:xtic(1) ls 1, \
'data.dat' every ::1::1 using 2:xtic(1) ls 2, \
'data.dat' every ::1::1 using 3:xtic(1) ls 2, \
'data.dat' every ::1::1 using 4:xtic(1) ls 2, \
'data.dat' every ::1::1 using 5:xtic(1) ls 2, \
'data.dat' every ::1::1 using 6:xtic(1) ls 2, \
'data.dat' every ::1::1 using 7:xtic(1) ls 2, \
'data.dat' every ::1::1 using 8:xtic(1) ls 2, \
'data.dat' every ::1::1 using 9:xtic(1) ls 2, \
'data.dat' every ::1::1 using 10:xtic(1) ls 2, \
'data.dat' every ::1::1 using 11:xtic(1) ls 2, \
'data.dat' every ::1::1 using 12:xtic(1) ls 2, \
'data.dat' every ::2::2 using 2:xtic(1) ls 3, \
'data.dat' every ::2::2 using 3:xtic(1) ls 3, \
'data.dat' every ::2::2 using 4:xtic(1) ls 3, \
'data.dat' every ::2::2 using 5:xtic(1) ls 3, \
'data.dat' every ::2::2 using 6:xtic(1) ls 3, \
'data.dat' every ::2::2 using 7:xtic(1) ls 3, \
'data.dat' every ::2::2 using 8:xtic(1) ls 3, \
'data.dat' every ::2::2 using 9:xtic(1) ls 3, \
'data.dat' every ::2::2 using 10:xtic(1) ls 3, \
'data.dat' every ::2::2 using 11:xtic(1) ls 3, \
'data.dat' every ::2::2 using 12:xtic(1) ls 3
Which gives me the colours, but the group names are gone and the groups are no longer separated?
Is there an easier way to get what I want? It is not problem to re-organise the data file.
Thanks in advance,
J.

[Gnuplot-info] problem controlling fit length within given x values

[smms <sm...@ho...>]

I am trying to fit my data and control their lengths between b x_min x_max.
So renamed my functions with corresponding x_min and x_max values. But not
succeeding to control their length. Any suggestion pls? Many thanks and here
are my data and codes.

x1	           y1	      sd1	   x2	    y2	          sd2	      x3	   y3	         
sd3
611.75	 44.330	2.015	   638.65	41.417	1.848	    627.4	45.311	4.668
640.65	 42.821	2.161	   619.10	47.912	1.176	    607.0	56.682	2.234
637.35	 49.475	2.672	   623.95	50.669	3.419	    584.5	62.197	2.266
617.00	 52.755	2.917	   602.20	55.996	1.508	    522.0	70.550	4.751
602.30	 59.032	3.325	   530.95	71.744	3.519	    413.5	80.528	4.412
545.90	 67.169	3.351	   436.20	80.660	2.877	    286.9	87.453	4.542
467.40	 84.709	5.217	   372.35	84.351	3.640	    209.3	89.979	2.185
336.50	 90.083	4.053	   249.50	96.568	3.355	    96.8	96.859	3.476

f8(x) = a8*x*x+b8*x+c8
f_p(x, min, max) = (x >= min && x <= max) ? f8(x) : 1/0
fit f_p(x,min,max) 'data.txt' u 1:2 via a8,b8,c8
plot 'data.txt' u 1:2:3 w errorbars t '800', f_p(x,min,max) t 'f(800)' 

min=250 ; max=624;	
f10(x) = a10*x*x+b10*x+c10
f_q(x, min, max) = (x >= min && x <= max) ? f10(x) : 1/0
fit f_q(x,min,max) 'data.txt' u 4:5 via a8,b8,c8
replot 'data.txt' u 4:5:6 w errorbars t '1000', f_q(x,min,max) t 'f(1000)'

min=97 ; max=585;	
f12(x) = a12*x*x+b12*x+c12
f_q(x, min, max) = (x >= min && x <= max) ? f12(x) : 1/0
fit f_q(x,min,max) 'data.txt' u 7:8 via a12,b12,c12
replot 'data.txt' u 7:8:9 w errorbars t '1200', f_q(x,min,max) t 'f(1200)'









--
View this message in context: http://gnuplot.10905.n7.nabble.com/problem-controlling-fit-length-within-given-x-values-tp17882.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] unexpected difference between plot and splot of binary files with unstructured data

[cédric o. <ced...@gm...>]

Hi all,

For those interested in plotting unstructured meshes with gnuplot from
*binary* files, I found a trick to solve the issue I submitted in a
previous post.

I used to do

> plot 'file.dat' binary record=(15,-1) format='%float' u 1:2:3 w l palette

where the the binary file file.dat was written using octave
(fwrite(file,vector,kind)).

The 15 is because I plot cells of 5 points of two coordinates + the
value at the point (we can do cell or vertex centered as wished).
The -1 tells gnuplot to read all records.

A cell has thus the structure
 cell=[ptA ptB ptC ptD ptA]  #ptA is repeated to close the cell

And I used octave to write the final 1D array

 vector=[ptAx ptAy ptAval ptBx ptBy ptBval ptCx ptCy ptCval  ptDx ptDy
ptDval  ptAx ptAy ptAval ..];

The problem was that gnuplot does know that we are passing from one
element to another one and can draw lines between the last point of an
element to the first of the next one in the file. This may be
undesirable if for instance two elements are well separated.

To prevent gnuplot from such linking and to reproduce the effect of a
blank record in ascii file, I have added two copies of an existing
first point to the definition of the cells

 cell=[ptA ptA ptB ptC ptD ptA ptA]

Adding the first and last instances of PtA now allows one to attribute
some like of a missing value to these 'virtual' points so that the
vector containing data reads

vector=[ptAx ptAy ptAmissingvalue ptAx ptAy ptAval ptBx ptBy ptBval
ptCx ptCy ptCval  ptDx ptDy ptDval  ptAx ptAy ptAval ptAx ptAy
ptAmissingvalue..];

Now I can use something like that using the gnuplot ?: operator and
the NaN value 1/0.

plot 'tmpbin.dat'  binary record=(21,-1) format='%float' u
1:2:($3<missingvalue?$3:1/0) w l pal

So that elements are not connected to each other now.

The file is a bit bigger but the time to write much decreased.

Hope this helps

cedric


> The plot command gives expected results (I would expect that I have
> used an ascii filename). However, the splot command does not as some
> cells of the unstructured mesh are now connected although they should
> not.
>

Re: [Gnuplot-info] Probelm in guessing logistic growth parameters for fitting

[smms <sm...@ho...>]

Thanks Hans for your reply. I had to scale Y for data quantification. 

Date: Fri, 6 Dec 2013 13:48:00 -0800
From: ml-...@n7...
To: sm...@ho...
Subject: Re: Probelm in guessing logistic growth parameters for fitting



	On 06.12.2013 21:00, smms wrote:


> So I would appreciate

> if someone could refer me to a good reference of how to make good guessing

> of initial parameters.


That's evidently impossible.  If it there were such a thing as a 

reference manual on how to do it, it would not be called "guessing".


Initial values have to come from your own understanding of the problem. 

  There is no magic bullet.


It says in "help fit" that "Non-linear least-squares fitting is an art!" 

  You're now learning just how true that statement is.


Anyway, your problem is not with the initial values, it's with the 

scaling of the parameters themselves.  You've missed an important lesson 

from "help fit" there.


> a = 1e-7

> b = 10

> k = 0.5

> g(x) = a/(1+b*exp(-k*x))


> fit g(x) 'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3) via a,b,k

> plot g(x),'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3):($3/800**3) w

> errorbars not


Hmmm... so why rescale your y axis so heavily, if all that achieves is 

giving your "a" paramter a magnitude that's wildly different from the 

others?


a = 50.0

fit g(x) 'ss_nici_raw_processed_201.txt' u (1/$1):2:3 via a,b,k


would have been equivalent, quicker to type, and with the additional 

benefit of actually using your y errors for the fit.


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

Sponsored by Intel(R) XDK 

Develop, test and display web and hybrid apps with a single code base.

Download it for free now!

http://pubads.g.doubleclick.net/gampad/clk?id=111408631&iu=/4140/ostg.clktrk
_______________________________________________

gnuplot-info mailing list

[hidden email]

Membership management via: https://lists.sourceforge.net/lists/listinfo/gnuplot-info


	
	
	
	

	

	
	
		If you reply to this email, your message will be added to the discussion below:
		http://gnuplot.10905.n7.nabble.com/Probelm-in-guessing-logistic-growth-parameters-for-fitting-tp17874p17876.html
	
	
		
		To unsubscribe from Probelm in guessing logistic growth parameters for fitting, click here.

		NAML
	 		 	   		  



--
View this message in context: http://gnuplot.10905.n7.nabble.com/Probelm-in-guessing-logistic-growth-parameters-for-fitting-tp17874p17880.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Which Symbol font in Postscript terminal?

[Nunzio L. <nun...@gm...>]

Hi all,

I am wondering which Symbol font the Postscript terminal is using. I am
producing .eps files and importing them in MS Word for Mac (the journal I'm
submitting my paper to only accepts manuscripts in MS Word) and the Symbol
characters sometimes disappear when the file is printed in PDF during the
submission procedure. Thus I think the only option to get always a good
result is to embed the fonts in the file through "fontfile add", which is
what I currently do with other fonts. But how do I know which fonts in my
system are called by the "Symbol" alias in  gnuplot?
Can you help me with that?

Regards,

N.

Re: [Gnuplot-info] Probelm in guessing logistic growth parameters for fitting

[Hans-Bernhard B. <HBB...@t-...>]

On 06.12.2013 21:00, smms wrote:

> So I would appreciate
> if someone could refer me to a good reference of how to make good guessing
> of initial parameters.

That's evidently impossible.  If it there were such a thing as a 
reference manual on how to do it, it would not be called "guessing".

Initial values have to come from your own understanding of the problem. 
  There is no magic bullet.

It says in "help fit" that "Non-linear least-squares fitting is an art!" 
  You're now learning just how true that statement is.

Anyway, your problem is not with the initial values, it's with the 
scaling of the parameters themselves.  You've missed an important lesson 
from "help fit" there.

> a = 1e-7
> b = 10
> k = 0.5
> g(x) = a/(1+b*exp(-k*x))

> fit g(x) 'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3) via a,b,k
> plot g(x),'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3):($3/800**3) w
> errorbars not

Hmmm... so why rescale your y axis so heavily, if all that achieves is 
giving your "a" paramter a magnitude that's wildly different from the 
others?

a = 50.0
fit g(x) 'ss_nici_raw_processed_201.txt' u (1/$1):2:3 via a,b,k

would have been equivalent, quicker to type, and with the additional 
benefit of actually using your y errors for the fit.

Re: [Gnuplot-info] gnuplot error ';' expected

[Hans-Bernhard B. <HBB...@t-...>]

On 06.12.2013 09:03, Dang, Siaufung wrote:

> plot [0:1] 'predstr.out' u 1:3 t "predicted" w p 2, \
                                                  ^^^^

The ^^^^ is where the problem is.

> I believe this is an issue related to the version of
> gnuplot I am using now, because at the time ATAT was developed, they
> were using older versions of it.

"Older versions" doesn't quite cover it.  That syntax ("linetype" option 
without the "linetype" keyword) has been discontinued ages ago.

See "help plot" to learn what the syntax is these days, then get your 
program to add "lt", "pt" and related option keywords before the numbers.

[Gnuplot-info] Probelm in guessing logistic growth parameters for fitting

[smms <sm...@ho...>]

I am having difficulty in guessing initial parameters values for logistic fit
plotting to my data set. The parameters I used actually from somebody else.
Going through gnuplot helpf fit did not help at all. So I would appreciate
if someone could refer me to a good reference of how to make good guessing
of initial parameters. Also if you happen to find a good values for a nice
fit, can you please explain your basis of choosing different parameters?
Included below are graph, data and codes.

This initial values are getting very frustrating and hence many thanks for
your any feedback in advance. 

<http://gnuplot.10905.n7.nabble.com/file/n17874/lgfit.png> 
x       y       sd
2.50	9.04	0.03
2.25	9.06	0.05
1.11	9.12	0.19
0.71	9.97	0.18
0.60	11.36	0.24
0.56	12.44	0.55
0.50	14.23	0.37
0.43	16.93	1.20
0.38	19.18	1.12
0.32	24.83	2.26
0.30	28.87	1.39
0.25	34.23	2.07
0.21	39.52	0.53
0.16	44.10	1.81
0.13	49.73	1.19

a = 1e-7
b = 10
k = 0.5
g(x) = a/(1+b*exp(-k*x))
fit g(x) 'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3) via a,b,k
plot g(x),'ss_nici_raw_processed_201.txt' u (1/$1):($2/800**3):($3/800**3) w
errorbars not





--
View this message in context: http://gnuplot.10905.n7.nabble.com/Probelm-in-guessing-logistic-growth-parameters-for-fitting-tp17874.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.