hi...i've been using mayavi for several days and can't seem to find a
way to do the following: *tick marks on almost any plot--can't find a single example on web with them *how to do surfaces over irregularly spaced data points...i have a hunch this should be something along the lines of scalar_scatter, then delauny2D, etc.....but so far haven't even gotten the first one to work... brandon ps: i like mayavi overall, it just needs more doc examples...especially non-flow-visualization ones... _______________________________________________ Enthought-Dev mailing list [hidden email] https://mail.enthought.com/mailman/listinfo/enthought-dev |
Hello,
> *tick marks on almost any plot--can't find a single example on web with them Have you tried mlab.axes() ? Or you mean you want tick marks on those axes? (If the latter, I don't know how to do it...) > *how to do surfaces over irregularly spaced data points...i have a > hunch this should be something along the lines of scalar_scatter, then > delauny2D, etc.....but so far haven't even gotten the first one to > work... Unfortunately, my experience has been that this is a highly data-dependent problem. I assume your problem here is that you have a bunch of data points with associated scalar values (e.g. an irregularly sampled 3D scalar field), and you want constant value surfaces, is that correct? One way you can do this is below... I'm generating fake data that is just a gaussian field sampled on an irregular grid: >>> from enthought.mayavi import mlab >>> from numpy.random import randn >>> from numpy import exp >>> x,y,z = randn(3,1000) >>> scalar = exp(-x*x-y*y-z*z) >>> ss = mlab.pipeline.scalar_scatter(x,y,z,scalar) >>> del3d = mlab.pipeline.delaunay3d(ss) You can now visualize the delaunay triangulartion. Say for example, a scalar cut plane : >>> mlab.pipeline.scalar_cut_plane(del3d) Or alternatively, an isosurface works decently for this case: >>> iso = mlab.pipeline.iso_surface(del3d) >>> iso.contour.auto_contours = True >>> iso.contour.number_of_contours = 3 >>> iso.actor.property.opacity = .5 >>> iso.contour.minimum_contour = .2 >>> iso.contour.maximum_contour = .7 Now if your data is too irregular or otherwise weird to get a good Delaunay triangulation (almost always the case for when I've done this), you can re-sample the data into a regular grid and start using volume-visualization techniques with mlab.pipeline.scalar_field, which IMHO are often more robust and useful. You can do this with mlab.pipeline.gaussian_splatter if the data volume isn't too large, but for much of what I've worked with, it's better to just write your own code to resample and then just load the resampled field directly... _______________________________________________ Enthought-Dev mailing list [hidden email] https://mail.enthought.com/mailman/listinfo/enthought-dev |
In reply to this post by j brandon keith
On Thu, Apr 01, 2010 at 11:18:24AM -0700, j brandon keith wrote:
> hi...i've been using mayavi for several days and can't seem to find a > way to do the following: > *tick marks on almost any plot--can't find a single example on web with them I am not sure what you want to do. Most of the 'ticking' functionnality is provided by the axes module, that you can access using mlab.axes http://code.enthought.com/projects/mayavi/docs/development/html/mayavi/auto/mlab_other_functions.html#axes > *how to do surfaces over irregularly spaced data points...i have a > hunch this should be something along the lines of scalar_scatter, then > delauny2D, etc.....but so far haven't even gotten the first one to > work... Hopefuly Erik gave you a good head start. However, if what you are trying to is more to 'connect the dots on a surface' than to 'iso-surface aftering interpolating between points', you might find the following example helpful: http://code.enthought.com/projects/mayavi/docs/development/html/mayavi/auto/example_surface_from_irregular_data.html HTH, GaĆ«l _______________________________________________ Enthought-Dev mailing list [hidden email] https://mail.enthought.com/mailman/listinfo/enthought-dev |
Free forum by Nabble | Edit this page |