numpy to matlab interface with mlabwrap

Question

I am looking for a simple way to visualize some of my data in numpy, and I discovered the mlabwrap package which looks really promising. I am trying to create a simple plot with the ability to be updated as the data changes.

Here is the matlab code that I am trying to duplicate

>> h = plot([1,2,3], [1,2,3], '-o');
>> set(h, 'XData', [0,0,0]);
>> drawnow();

to python

>> from mlabwrap import mlab
>> h = mlab.plot([1,2,3], [1,2,3], '-o')
>> mlab.set(h, 'XData', [0,0,0])
>> mlab.drawnow();

However, the second to last command fails with an error message

error: One or more output arguments not assigned during call to "set".

Any suggestions on how to fix this?

Concavoconvex

Answer 1

Maybe mlab is mad that you're not saving matlab's return value for that set() call...

I don't have this installed, what does someval = mlab.set(h,'XData') give?

edit: you could also try using nout... mlab.set(h,'XData',[0,0,0],nout=0)

Jessjessa

Thanks, you bet me to the answer by a few minutes. By default, mlabwrap translates the python command to [RES0__]=set(arg0__, arg1__, arg2__);, and since set takes no ouptut arguments, this results in the error above. – Concavoconvex

Answer 2

Since set takes no output arguments, we need to tell mlabwrap that no output arguments should be given to it to avoid the error message above.

mlab.set(h, 'XData', [0,0,0], nout=0)

Concavoconvex