Gnuplot Density Mapping: Tips and Tricks for Efficient Data Plotting

[s_hy]

Hi all,

I used below bash command to plot my data (.dat file)

#!/bin/bash

(
echo 'set term jpeg'
echo 'set style data lines'
echo 'set yrange [0:200]'
echo 'set xrange[0:200]' 
echo 'set pm3d map'
echo 'set palette defined (-2 "yellow", 0 "green", 2 "red")'for f in "$@"
do
#	echo "Processing $f" >&2
	pic_name="${f%.dat}.jpg"
	echo "set out '$pic_name'"
	echo "splot '$f'"
done) | gnuplot

but I got this warning:

Warning: Single isoline (scan) is not enough for a pm3d plot.
Hint: Missing blank lines in the data file? See 'help pm3d' and FAQ.

As I know, to map using gnuplot for 3d density mapping, I should have output with blank lines to separate different x values, like this

-180.0 -180.0 0.0
-180.0 -179.0 0.2
[...]
-180.0  180.0 0.0

-179.0 -180.0 0.4
-179.0 -179.0 0.2
[...]
-179.0  180.0 0.0

[...]

but, if I have a few thousand lines, with 200 .dat files, what can I do to expedite my job? Anyone can help...thank you in advance.

 

[chiro]

Hey s_hy and welcome to the forums.

One idea I have for you is to run a script (or write some code to create an executable) to just remove every instance of a double or more than double CR-LF (carriage return line-feed) and just set it to a single CR-LF.

 

[s_hy]

chiro,

how can I do that? gnuplot is new to me..still not familiar with the command

 

[chiro]

It's not a gnuplot specific suggestion: it's a general one that can be implemented on your language (scripting/non-script/whatever) and platform of your choice.

All you want to do is basically read in every line of the file that is relevant and whenever you encounter a file with multiple CR-LF (carriage return-line feed) values just remove all of them but one.

If you want to know what these values are use a hex-editor and look at the raw data and see if it's just a one or two byte code and you'll see what the values are.

The simplest way to do this without even worring about the CR-LF values is to just have a loop that reads in one line at a time and then if the line is blank forget about it and if not write it to another file.

When the loop terminates you'll have another file without any extra spaces in the data and you can use that.

It's basically just an open statement, one loop that reads until end of file, read line, check if empty or with only spaces and if so don't add it to output but if not write that line to the output file, loop terminates, close input file, flush output to existing file close handle, program exits and you have a file that has no gaps.

If you have any platform that allows you to read in one line at a time and write one line at a time you're looking at something like 10-20 lines for the entire thing and will do its magic pretty much on the spot.

 

[LudusRex]

Hello,

Thank you for sharing your experience with using gnuplot for density mapping. It seems like you have encountered a common issue with plotting data using gnuplot - the need for blank lines to separate different x values. This is necessary for creating a proper 3D density map.

One solution to expedite your job would be to use a script or program to automatically add the necessary blank lines to your data files. This can save you time and effort, especially if you have a large number of files to plot.

Another tip for efficient data plotting is to use the "plot for" command in gnuplot, which allows you to plot multiple data files at once. This can save you from having to manually input each file name in your script.

Additionally, you may want to consider using a different plotting software that is specifically designed for 3D density mapping. There are many options available, such as ParaView, VisIt, and Tecplot, which may offer more efficient and user-friendly tools for your specific needs.

I hope these tips and suggestions are helpful for your data plotting. Good luck with your research!