next up previous index
Next: Octave With Grace Up: Drawing Graphs Previous: GNUPlot   Index

Click for printer friendely version of this HowTo

Octave

One thing to note about using Octave, is that to generate graphs, it uses GNUPlot, so a lot of the commands are the same.

Here is how to create one of the normal curve graphs that are so common in the paper:

> x = linspace(-8, 8, 100);
> y = normal_pdf (x);
> plot(x, y)

If we wanted to replace the ``line 1'' label that is the default, we could change our plot command to add our own version:

> plot(x, y, ``;mu = 0, sigma = 1;'')

To export the graph as a color EPS file, all you need to do is:

> gset term postscript eps color;
> gset output "normal_curve.eps"
> replot

Sometimes when you are using octave, you may want to export the data that you are working with so you can import it into another graphing program other than gnuplot (for example, you may want to import the data into xmgrace). To do this requires a little bit of a trick since you'll notice that octave exports vectors as a single line with a whitespace between the elements. While this makes perfect sense to do, it is often the case that you have a vector representing the values for the x-axis and a vector (or vectors) representing values on the y-axis (as we do in the example we just gave). If we were to export these vectors directly, we'd get the values for the x and the y axis on two different lines and this is not what we want. Instead, we'd like to have each line in our output file correspond to a value from the x-axis vector and a value from the y-axis vector (or values from y-axis vectors). To do this, you simply create a new matrix that is made up of the transposes of these vectors. To continue with the previous example, if we wanted to export our data in a format that could easily be imported into xmgrace we would do this:

> x = linspace(-8, 8, 100);
> y = normal_pdf (x);
> z = [x', y'];
> save -ascii ``output_file.txt'' z


next up previous index
Next: Octave With Grace Up: Drawing Graphs Previous: GNUPlot   Index

Click for printer friendely version of this HowTo

Frank Starmer 2004-05-19
>