gnuplot basics

7/30/2019 Gnuplot Basics

1/5

gnuplot in Action, Philipp K Janert, 2010 by Manning Publications Co.

This document's conventions: gnuplot features are going to be bulleted and

mathematical/data analysis theory without bullet.

Part 1: Basics

A tool for visualizing data and analyzing it using graphical methods.

Chapter 1: Prelude

The art of discovering relationships in data and extracting information from it by visual

means is called graphical analysis. gnuplot reads data from simple text files, with the data arranged in columns.

The plotcommand requires the name of the data file as argument in quotes. The

rest of the command line specifies which columns to use for the plot, and in which

way to represent the data.

The using 1:2declaration tells gnuplot to use the first and second column in the

file called marathon. The final part of the command, with boxes, selects a box

style, which is often suitable to display counts of events.

plot "marathon" using 1:2 with boxes

gnuplot handles most everything else by itself: it sizes the graph and selects the

most interesting plot range, it draws the border, and it draws the tic marks and

their labels. All these details can be customized.

Diffusion limited aggregation (DLA), a process wherein a particle performs a random

walk until it comes in contact with a growing cluster of particles. At the moment of

contact, the particle sticks to the cluster at the location where the contact occurred and

becomes part of the cluster. Now, a new random walker is released to perform a

random walk, until it sticks to the cluster. And so on. DLA clusters are fractals. If you

want to learn more about DLA and similar processes, check out Fractals, Scaling, and

Growth Far From Equilibrium by Paul Meakin (1998).

Setting axes labels and logscales: In gnuplot, all details of a graphs appearance

are handled by setting the appropriate options. To place the labels on the x and y

axes in figure 1.3, I used

set xlabel "Cluster Size"

set ylabel "Run Time [sec]"

Figure 1.4 is drawn using double-logarithmic axes. This is another option, which is

set as follows:


2/5

set logscale

Plotting Several mathematical functions together: we just list them one after

another on the command line for the plot command:

plot "runtime" using 1:2 title "data" with lines,

1.2*(x/1000)**2.7 title "model"

title directive takes a string as argument, which will be displayed together with

a line sample in the plots key or legend.

Iteration: a crucial aspect of graphical analysis: plotting the data this way and that way;

comparing it to mathematical functions or to other data sets; zooming in on interesting

regions or zooming out to detect the overall trend; applying logarithms or other data

transformations to change its shape; using a smoothing algorithm to tame a noisy data

set; and so on. Ingenuity, intuition, and curiosity are the most important character traits.

Everyone can play this game, if theyre interested in finding out what the data tries totell them.

Chapter 2: Essential gnuplot

If gnuplot is installed on your system, it can usually be invoked by issuing the

command:

gnuplot

Once launched, gnuplot displays a welcome message and then replaces the shellprompt with a gnuplot> prompt. Anything entered at this prompt will be interpreted

as gnuplot commands until you issue an exit or quit command, or type an

end-of-file (EOF) character, usually by hitting Control-D.

define the plot range that we are interested in:

plot [][-2:2] sin(x), x, x-(x**3)/6

The range is given in square brackets immediately after the plot command. The first

pair of brackets defines the range of x values (we leave it empty, since were

happy with the defaults in this case); the second restricts the range of y valuesshown.

Plot of a series of sine functions looked like

sin(x/4), sin(x/2), sin(x), sin(2*), sin(4*x)


3/5

Plotting from a File

While whitespace-separated numerical data is what gnuplot expects natively, recent

versions of gnuplot can parse and interpret significant deviations from this norm, including

text columns (with embedded whitespace if enclosed in double quotes), missing data,

and a variety of textual representations.

Assuming that the file is called prices, we can simply type

plot "prices"

Since data files typically contain many different data sets, well usually want to

select the columns to be used as x and y values.

This is done through the usingdirective to the plot command:

plot "prices" using 1:2

This will plot the price of PQR shares as a function of time: the first argument to

the usingdirective specifies the column in the input file to be plotted along the

horizontal (x) axis, while the second argument specifies the column for the

vertical (y) axis.

If we want to plot the price of XYZ shares in the same plot

plot "" using 1:2, "" using 1:3

By default, data points from a file are plotted using unconnected symbols. we

need to tell gnuplot what style to use for the data. This is done using the with

directive. Many different styles are available. Among the most useful ones are

with linespoints, which plots each data point as a symbol and also connects


4/5

subsequent points, and with lines, which just plots the connecting lines, omitting

the individual symbols.

We can do much better by including a title for each data set as part of the plot

command:

plot "prices" using 1:2 title "PQR" with lines,

The title has to come after the using directive in the plot command.

If its missing entirely and the data file contains multiple columns, gnuplot plots

the second column versus the first (this is equivalent to using1:2). If a using

directive is given, but lists only a single column, gnuplot will use this column for

y values and provide x values as integers starting at zero. This is also what

happens when no using is given and the data file contains only a single

column.

Saving and exporting: Gnuplot commands can be saved to a file simply using the

save command:

save "graph.gp"

This will save the current values of all options, as well as the most recent plot

command, to the specified file. This file can later be loaded again using the load

command:load "graph.gp"

The effect of loading a file is the same as issuing all the contained commands.


5/5

Call command: An alternative to loadis the callcommand, which is similar to load,

but also takes up to 10 additional parameters after the filename to load. The

parameters are available inside the loaded file in the variables $0 through $9.

Strings are passed without their enclosing quotes, and the special variable $#

holds the number of parameters to call. We can use call to write some simple

scripts for gnuplot.

If the commands shown in the listing are placed into a file, this file can be

executed using call and will write the most recent plot to a PNG file and

restore the initial session to its original state:

set terminal push #save the current terminal settings

set terminal png #change terminal to PNG

set output "$0" #set the output filename to the first option

replot #repeat the most recent plot command

set output #restore output to interactive mode

set terminal pop #restore the terminal

Push and Pop Pseudo-terminals: weve used the two pseudo-terminals push

and pop to help with the back-and-forth between the interactive and file

terminals. The former (push) saves the current terminal settings onto a stack; the

latter (pop) restores the terminal to the state saved on the stack. Neither makes

any assumptions about the choice of interactive terminal, and therefore both

can safely be used in scripts that must be portable across architectures. Process of Plotting: Assuming the file shown in the listing was named export.gp

in the current directory, we would call it like this, to write the current plot to a

file called graph.png:

call "export.gp" "graph.png"

Here, both arguments are quoted. Quoting the second argument isnt strictly

necessary, but highly recommended to avoid unexpected parsing of the output

filename. The quotes are stripped before the argument is made available to the

command file in the $0 variable.

Exporting the graph to a file:

plot exp(-x**2)

set terminal png

set output "graph.png"

replot

Always save the commands used to generate a plot to a command file before

exporting to a printable format. This can only be done from the commands savedto file using save, not from plots exported to a graphics file.

gnuplot basics

Documents