gnuplot-info — For questions and discussion about gnuplot

[Gnuplot-info] Projection of 2D color graph on X axis

[1sezgin <1se...@gm...>]

I have a problem with gnuplot. ( Version 4.2 patchlevel 2)
With the code which I past belove I produce a 2D color image. I would
like to make projection on x axis which mean by using a data like;
x    y    z
1    1    2
1    2    0
1    3    1
1    4    1
2    1    1
2    2    0
2    3    1
2    4    0
3    1    2
3    2    0
3    3    1
3    4    2

as a result of projection I want to draw these new lines  (it should
sum z values for every y group (from 1 to 4)

x   new y
1      4
2      2
3      5

is it possible to make this with gnuplot?

My goal is to draw one divided graph, which has a color graph up
(first  data) and projection of first data (second data) like a
histogram, down.
Sezgin

set xlabel 'A1'
set ylabel 'B1'
set view map
#set palette rgbformulae 10,13,33
set palette gray
set cbrange [0:5]
set xrange [0:3276]
set yrange [-1638:1638]
set term png size 840,840
set output 'image.png'

splot 'ASCIIdata.txt' using (0.7*($1+$2)):(0.7*($1-$2)):3 notitle with
points palette ps 1 pt 5

-- 
View this message in context: http://www.nabble.com/Plot-an-ellipse-knowing-center-and-axis-tp20184644p20187190.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] fit function weirdness

[peterchun <pet...@ma...>]

Hi,
I am having some problems with the fit of a straight line.

I want to get a line of best fit for the 3rd column.
It really should give 
for 
f(x)=m*x+c
m=0
c=293255

but I get 
m=close to 0
c=1

I do not understand how I get a line of best fit that does not fit
What am I doing wrong?

Thanks
Peter


data:
26/08/08 293449 293255
10/09/08 296787 293255
12/09/08 296787 293255
15/09/08 296787 293255
17/09/08 296787 293255
18/09/08 296787 293255
19/09/08 296787 293255
20/09/08 296787 293255
21/09/08 296787 293255
22/09/08 296787 293255
23/09/08 296787 293255
24/09/08 296787 293255
25/09/08 296787 293255
26/09/08 296787 293255
27/09/08 296787 293255
28/09/08 296787 293255
29/09/08 296787 293255
30/09/08 296787 293255
01/10/08 296787 293255
02/10/08 296787 293255
03/10/08 296787 293255
04/10/08 296787 293255
05/10/08 296787 293255
06/10/08 296787 293255
07/10/08 296787 293255
08/10/08 296787 293255
09/10/08 296787 293255
10/10/08 296787 293255
11/10/08 296787 293255
12/10/08 296787 293255
13/10/08 296787 293255
14/10/08 296787 293255
15/10/08 296787 293255
16/10/08 296787 293255
17/10/08 296787 293255
18/10/08 296787 293255
19/10/08 296787 293255
20/10/08 296787 293255
21/10/08 296787 293255
22/10/08 296787 293255
23/10/08 296787 293255
24/10/08 296787 293255
25/10/08 296787 293255
26/10/08 296787 293255
27/10/08 296787 293255
28/10/08 296787 293255
29/10/08 296787 293255



gnuplot commands:
dataset="data"
set xdata time
set timefmt "%d/%m/%y"
set style data boxes
set style fill solid 1 noborder
set xrange [ "29/07/08":"29/01/09" ]
set title "data"
set xlabel "date"
set ylabel "whatever"
set grid
f(x)=m*x + c
fit [ "29/09/08":"28/10/08" ] f(x) dataset using 1:3 via m,c
plot dataset using 1:2 title "line1", dataset using 1:3 title "line2" , f(x)
title "Trendline" lw 3



-- 
View this message in context: http://www.nabble.com/fit-function-weirdness-tp20218720p20218720.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Do a Gnuplot of a file in python

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

Santix wrote:

> g.load(power.p)
> 
> but it gives me this error:
> 
> Traceback (most recent call last):
>   File "./spectrum_output.py", line 310, in <module>
>     main_loop(tb)
>   File "./spectrum_output.py", line 289, in main_loop
>     g.load(power.p)		
> AttributeError: 'file' object has no attribute 'p'

I'm reasonably sure that this has absolutely nothing to do with gnuplot. 
  It may have something to do with gnuplot.py --- but that's not 
produced by us.  Or it may be a plain an simple mistake in how you used 
Python.

So I can only offer a wild guess: try putting "" around the filename, i.e.:

	g.load("power.p")

Re: [Gnuplot-info] BREAK: Singular matrix in Invert_RtR

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

ray.zong wrote:
> BREAK:          Singular matrix in Invert_RtR
[...]
> Anyone knows the meaning of that message.

The message means that your fit failed.  The determination of new 
parameters failed because the system of equations became invalid.

There are several possible reasons for this, but the most probable ones 
are that your fit parameters are mutually dependent, or that at least 
one of them just doesn't have any effect on the fit's quality.  That's 
all that can be said without seeing the actual commands and data you used.

[Gnuplot-info] 2D color image projection

[1sezgin <1se...@gm...>]

I have a problem with gnuplot. ( Version 4.2 patchlevel 2 )
With the code which I past belove I produce a 2D color image. I would
like to make projection on x axis which mean by using a data like;
x    y    z
1    1    2
1    2    0
1    3    1
1    4    1
2    1    1
2    2    0
2    3    1
2    4    0
3    1    2
3    2    0
3    3    1
3    4    2

as a result of projection I want to draw these new lines  (it should
sum z values for every y group (from 1 to 4)

x   new y
1      4
2      2
3      5

is it possible to make this with gnuplot?

My goal is to draw one divided graph, which has a color graph up
(first  data) and projection of first data (second data) like a
histogram, down.

Sezgin

set xlabel 'A1'
set ylabel 'B1'
set view map
#set palette rgbformulae 10,13,33
set palette gray
set cbrange [0:5]
set xrange [0:3276]
set yrange [-1638:1638]
set term png size 840,840
set output 'image.png'

splot 'ASCIIdata.txt' using (0.7*($1+$2)):(0.7*($1-$2)):3 notitle with
points palette ps 1 pt 5

-- 
View this message in context: http://www.nabble.com/2D-color-image-projection-tp20187336p20187336.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plot an ellipse knowing center and axis

[skuallpa <re...@gm...>]

Ok it works well know
thanks a lot

-- 
View this message in context: http://www.nabble.com/Plot-an-ellipse-knowing-center-and-axis-tp20184644p20189537.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] re mapping matrix plot axis

[Thomas S. <t.s...@fz...>]

you need the actual gnuplot version.
type 'help matrix ascii' or have a look at
http://www.gnuplot.info/docs/node332.html

-- 
View this message in context: http://www.nabble.com/remapping-matrix-plot-axis-tp20187248p20188524.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plot an ellipse knowing center and axis

[Thomas S. <t.s...@fz...>]

> I am new to gnuplot and I need to draw an ellipse on a distribution of
data. 
> I know the center of the ellipse and the semimajor and semiminor axis.
> 
> Has someone some example code ?

center of ellipse: xm,ym
semimajor axis: a
semiminor axis: b

for example:

xm = 5
ym = 2
a = 2
b = 1

set xrange [-1:10]
set yrange [-1:5]
set zeroaxis
set parametric

fx(t)=xm+a*cos(t)
fy(t)=ym+b*sin(t)

set arrow 1 from xm,ym to (xm+a),ym
set arrow 2 from xm,ym to xm,(ym+b)

plot fx(t),fy(t)

-- 
View this message in context: http://www.nabble.com/Plot-an-ellipse-knowing-center-and-axis-tp20184644p20187442.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Plot an ellipse knowing center and axis

[skuallpa <re...@gm...>]

Hello,

I am new to gnuplot and I need to draw an ellipse on a distribution of data. 
I know the center of the ellipse and the semimajor and semiminor axis.

Has someone some example code ?
Thanks in advance

-- 
View this message in context: http://www.nabble.com/Plot-an-ellipse-knowing-center-and-axis-tp20184644p20184644.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Error in curve when plotting on reverse logarithmic x-axis

[Maximilian S. <ste...@gm...>]

Thank you for your quick replys. It was quite easy.

Of course the line starts at (0,1), sorry for that. Because of the
logarithmic scale this point can't be printed like all other points with
x=0. So this is the reason why gnuplot act strange (in my opinion), it
starts at the point (1,0.32) which does't exist, but (0.005,0.32) is the
next existing point so 0.32 makes sense in some way. 
I agree that it would be nicer if gnuplot would ignore zero values on
logscales or print an error.

When I delete all x=0 values it works fine. The zero values have no
relevance for this chart either.

Thanks again for your quick replys.


P.S.: 0e-20 means 0 times 10 to the power of -20 (0*10^-20) so it's 0 (in
the db and in my python script it is zero, i don't know why it's exported
like that).

P.P.S.: 0 to the power of anything is 0. Except 0^0, that's 1 (per
definition)
-- 
View this message in context: http://www.nabble.com/Error-in-curve-when-plotting-on-reverse-logarithmic-x-axis-tp20131291p20141059.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Error in curve when plotting on reverse logarithmic x-axis

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

Maximilian Stein wrote:

> The line starts at point 0:1 (coordinate origin)

No.  It actually _ends_ there, and that's (1,0), not (0,1): x is 1, y is 0.

> and is printed correctly
> until the last point. But from there it returns horizontally to the y-axis
> to the point 0:<last printed y-coordinate>.

Where else did you expect it to go?  The x coordinate for the first 20 
data points or so is not only outside your given range, but completely 
invalid on a logarithmic axis.  Zero has no place on such an axis.

[Gnuplot-info] Do a Gnuplot

[Santix <san...@gm...>]

-- 
View this message in context: http://www.nabble.com/Do-a-Gnuplot-tp20133793p20133793.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Do a Gnuplot of a file in python

[Santix <san...@gm...>]

I am doing a python program that save the data in a text file in columns and
I want to do a gnuplot to plot the results.
But I want the program in python to show the result with gnuplot.
I have tried this:

g.load(power.p)

but it gives me this error:

Traceback (most recent call last):
  File "./spectrum_output.py", line 310, in <module>
    main_loop(tb)
  File "./spectrum_output.py", line 289, in main_loop
    g.load(power.p)		
AttributeError: 'file' object has no attribute 'p'

On power.p I have this:

set autoscale
unset logscale
unset label
set xtic auto
set ytic auto
set title "Spectrum Analyzer"
set xlabel "Frecuency"
set ylabel "Power (dB)"
plot "power.dat" using 1:2 title 'm.data[0]' with linespoints

And in power.dat:
2231500000.0    18.2582263947 
2234500000.0    17.0396003723
        ...                      ...
with more data

It goes ok when I load it in gnuplot, but not when I execute the python
program in the Terminal.

Thanks in advance!

Regards,

Santiago Ortega.


-- 
View this message in context: http://www.nabble.com/Do-a-Gnuplot-of-a-file-in-python-tp20133953p20133953.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Error in curve when plotting on reverse logarithmic x-axis

[Maximilian S. <ste...@gm...>]

Hi everybody,
I have to create a line chart with a reverse logarithmic x-axis starting at
1 becoming smaller (to 1e-4). I read the coordinate data from a database,
print the configuration and data to a file and then plot it with gnuplot.
The line starts at point 0:1 (coordinate origin) and is printed correctly
until the last point. But from there it returns horizontally to the y-axis
to the point 0:<last printed y-coordinate>.

The chart:
------------------------------------------------------
http://www.nabble.com/file/p20131291/not_correct.emf not_correct.emf 
------------------------------------------------------

Here's the file I use to plot the chart:
------------------------------------------------------

set terminal emf color dashed font "arial,12" size 800,600
set output "not_correct.emf"
set grid lt 28
set logscale x
set xrange [1e-4:1] reverse
set yrange [0:1]
set key box reverse Left top right spacing 1.5
plot '-' title "Data" with lines lt 1 lw 3
0E-20 1.00000000000000000000
0E-20 1.00000000000000000000
0E-20 0.52000000000000000000
0E-20 0.52000000000000000000
0E-20 0.52000000000000000000
0E-20 0.52000000000000000000
0E-20 0.50000000000000000000
0E-20 0.50000000000000000000
0E-20 0.50000000000000000000
0E-20 0.50000000000000000000
0E-20 0.50000000000000000000
0E-20 0.46000000000000000000
0E-20 0.44000000000000000000
0E-20 0.44000000000000000000
0E-20 0.44000000000000000000
0E-20 0.44000000000000000000
0E-20 0.42000000000000000000
0E-20 0.42000000000000000000
0E-20 0.42000000000000000000
0E-20 0.42000000000000000000
0E-20 0.42000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.40000000000000000000
0E-20 0.36000000000000000000
0E-20 0.36000000000000000000
0E-20 0.36000000000000000000
0E-20 0.34000000000000000000
0E-20 0.34000000000000000000
0E-20 0.34000000000000000000
0E-20 0.34000000000000000000
0E-20 0.34000000000000000000
0E-20 0.34000000000000000000
0.00500000000000000000 0.32000000000000000000
0.00500000000000000000 0.30000000000000000000
0.00500000000000000000 0.30000000000000000000
0.01000000000000000000 0.26000000000000000000
0.01000000000000000000 0.26000000000000000000
0.03000000000000000000 0.24000000000000000000
0.03000000000000000000 0.24000000000000000000
0.03000000000000000000 0.24000000000000000000
0.04000000000000000000 0.24000000000000000000
0.10000000000000000000 0.20000000000000000000
0.18000000000000000000 0.16000000000000000000
0.26000000000000000000 0.14000000000000000000
0.45500000000000000000 0.12000000000000000000
0.45500000000000000000 0.12000000000000000000
0.64000000000000000000 0.04000000000000000000
0.80500000000000000000 0.04000000000000000000
0.80500000000000000000 0.04000000000000000000
1.00000000000000000000 0E-20
e

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

I'm quite new to gnuplot and I have no idea how to fix this.

Any help is very appreciated
Maximilian Stein
-- 
View this message in context: http://www.nabble.com/Error-in-curve-when-plotting-on-reverse-logarithmic-x-axis-tp20131291p20131291.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

[Gnuplot-info] Need fixed relationship between Y1 and Y2

[Tai V. <ta...@ea...>]

Hi gnuplot gurus!

I'm graphing a value which is measured in different units in different
countries, but the units are easy to convert with a simple division by
293.071.
How can I arrange to plot accurately against Y1 in one unit and Y2 in the
other?

I have tried graphing the same data twice (divided by the constant in one
case), but the scaling is not exactly the same so the Y1 and Y2 tics do not
agree.

A second problem with this approach is that I need to make some of the
points invisible.

Thoughts? I feel like this should be something I can solve myself, but I
haven't figure out how yet. Pointers or suggestions very welcome.

tai viinikka

-- 
tai = ta...@ea... ::: We are moving as fast as we can
                   ::: across the dangerous surface of
                ::: the world. -Candas Jane Dorsey

[Gnuplot-info] time/date samples and binary data file

[Sylvain B. <syl...@gm...>]

Hi all,
Is it possible to provide time/date data to gnuplot from a binary data file?
The time data would be 32 bits or 64 bits time_ts.

regards,

 Sylvain

-- 
Code in C only, use GPL licensed only software, use GNU/Linux and
secure your digital freedom!

[Gnuplot-info] BREAK: Singular matrix in Invert_RtR

[ray.zong <zxr...@gm...>]

Hi everyone.

I was using gnuplot to fit some data points into an equation. However, I
encountered this message

BREAK:          Singular matrix in Invert_RtR

and no actual result shows.

Anyone knows the meaning of that message.

thanks.

xuanran
-- 
View this message in context: http://www.nabble.com/BREAK%3A----------Singular-matrix-in-Invert_RtR-tp20036467p20036467.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Problem with splot

[Thomas S. <t.s...@fz...>]

> Thanks for the help!
>
> Two more questions...
>
> Is it possible to get a z-scale where "-" is "up"?

set zrange [*:*] reverse

> Does anyone have a nice palette? I usually use the
> "standard" or "set palette defined (-3 "blue", 0 "white", 1 "red")"?

have a look at 'help palette' and then see what the actual
palette looks like with 'test palette'.
-- 
View this message in context: http://www.nabble.com/Problem-with-splot-tp19970779p20029493.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Problem with splot

[l_johan_k <l_j...@ho...>]

Thanks for the help!

Two more questions...

Is it possible to get a z-scale where "-" is "up"?

Does anyone have a nice palette? I usually use the "standard" or "set
palette defined (-3 "blue", 0 "white", 1 "red")"?

/Johan
-- 
View this message in context: http://www.nabble.com/Problem-with-splot-tp19970779p20009089.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Problem with splot

[Thomas S. <t.s...@fz...>]

> One more question...
>
> Since my graph consists of discrete points I would like to
> have a flat surface between the values (preferably in the
> xy-plane). Is this possible?
>
> (I.e. I would like to have a "cone" for each matrix element,
> but with a flat surface between the elements.)
>
> Its hard to explain...

you are looking for a  3d bar chart or lego plot.
well, that's not possible, AFAIK, but a poor man's
solution would be:

you may use impulses:

splot "filename.dat" matrix w imp lw 5

to have a surface at z=0:

set iso 6,6
splot "filename.dat" matrix w imp lw 5, 0

or

set iso 6,6
splot "filename.dat" matrix us 1:2:($3<0?$3:0) w imp lc 3 lw 10, 0, ""
matrix us 1:2:($3>0?$3:0) w imp lc 1 lw 10

-- 
View this message in context: http://www.nabble.com/Problem-with-splot-tp19970779p19999992.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plotting a time series (elapsed time) with time values > 24 hours

[Thomas S. <t.s...@fz...>]

> However, since I do generate the input datafile myself
> I could change to another format.
> Do you have any suggestions of a better way of doing this?
> I want to label the time axis in elapsed hours in a reasonably
> common format.
> I think "24:00" is more intuitive than "24.0" or "24".

you could write the time values as hours in floating point format:
e.g. '59.99139' instead of '59:59:29'

and then plot with:

set format x "%.0f:00"
set xtics 6
plot 'test.dat' w l

-- 
View this message in context: http://www.nabble.com/Plotting-a-time-series-%28elapsed-time%29-with-time-values-%3E-24-hours-tp19975946p19999320.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Problem with splot

[l_johan_k <l_j...@ho...>]

Thomas, many thanks! Now its working.

One more question...

Since my graph consists of discrete points I would like to have a flat
surface between the values (preferably in the xy-plane). Is this possible?

(I.e. I would like to have a "cone" for each matrix element, but with a flat
surface between the elements.)

Its hard to explain...

/Johan
-- 
View this message in context: http://www.nabble.com/Problem-with-splot-tp19970779p19998340.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plotting a time series (elapsed time) with time values > 24 hours

[Armin A. <aar...@nd...>]

Thomas Sefzick wrote:
> 
> BTW, most of your time values in the datafile, e.g. '59:59:29', are far
> from being valid times!
> '23:59:59' is the maximum value.
> 

Thanks Thomas, I think your suggestion with the explicit setting of xtics
works well.

To be honest I'm not very familiar with "proper" time formats. It worked in
Excel so I assumed it's okay.

However, since I do generate the input datafile myself I could change to
another format.
Do you have any suggestions of a better way of doing this? I want to label
the time axis in elapsed hours in a reasonably common format. I think
"24:00" is more intuitive than "24.0" or "24".


-- 
View this message in context: http://www.nabble.com/Plotting-a-time-series-%28elapsed-time%29-with-time-values-%3E-24-hours-tp19975946p19996824.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plotting a time series (elapsed time) with time values > 24 hours

[Thomas S. <t.s...@fz...>]

BTW, most of your time values in the datafile, e.g. '59:59:29', are far
from being valid times!
'23:59:59' is the maximum value.

values larger than '23:59:59' are accepted only because instead
of using the standard C function 'strptime' gnuplot has its own
one which fills the 'tm' structure and then calls 'mktime' which
fixes the wrong hour values.
-- 
View this message in context: http://www.nabble.com/Plotting-a-time-series-%28elapsed-time%29-with-time-values-%3E-24-hours-tp19975946p19996406.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.

Re: [Gnuplot-info] Plotting a time series (elapsed time) with time values > 24 hours

[Thomas S. <t.s...@fz...>]

> I know that this is not a common date/time format.
> In Excel I can specify [h]:mm as my formatting, in
> which case the hours keep incrementing beyond 24.
> I was just wondering if gnuplot has a similar functionality.

gnuplot uses the 'strftime' function from some C library,
and there are no 'accumulated' hours.

the easiest way to get ticslabels from '00:00' to '60:00'
is to define the xticslabels explicitly:

set xtics ("00:00" 0, "06:00" 6*60*60, "12:00" 12*60*60, "18:00" 18*60*60,
"24:00" 24*60*60, "30:00" 30*60*60, "36:00" 36*60*60, "42:00" 42*60*60,
"48:00" 48*60*60, "54:00" 54*60*60, "60:00" 60*60*60)

an explanation: your data start at '00:00:00', that means at 00:00
of the first day of the epoch (since you don't give a date), and
that's second '0'.
in the 'set xtics' command you can give the ticslabel position as
a date string or as an integer representing the number of seconds.

-- 
View this message in context: http://www.nabble.com/Plotting-a-time-series-%28elapsed-time%29-with-time-values-%3E-24-hours-tp19975946p19995644.html
Sent from the Gnuplot - User mailing list archive at Nabble.com.