Pynax

Minimal visualization toolkit for Python (Python n-axis).

Pynax:

• offers a user friendly interface to display your multidimensional data in Python through the matplotlib API
• brings native interactivity to your plots. Easy as pie !

Warnings

Pynax uses matplotib animated mode. Therefore, you must ensure that all your plotsa and matplotlib images have not been garbage collected to ensure interactivity. If at some point Pynax shows a non-interactive figure, this is most likely because one of your plot or image is no more in memory. This also implies that you have create brand new objects for each of your 'show'. You can't reuse previous objects and just 'draw' them.

Pynax uses axis related method instead of general methods. So, if you 'draw' a matrix, you cannot use pl.colorbar() to add a color bar to it. You have to specify image and axis (please see pynax.view.colorbar_mixin for an example of how to add a colorbar properly)

Pynax is dependant of some matplotlib mechanisms. For example, when displaying a 2D slice from multidimensional data, vmin and vmax are obviously calculated on the slice. If you change the values of the displayed slice (say, for example, that you have 3D data and that you show another slice), vmin and vmax will not be recomputed. A good practice is to always define vmin and vmax in the display options when plotting multidimensional data (this will become mandatory in a near future).

Concepts

Pynax uses some concepts above matplotlib ones (but close to) to represent plots.

• Mark: a mark is simply a value shared between different plots. If, at some point, a mark is updated, then its value will be updated in all the plot where it appears. A mark has some intrisic properties, like color, that are used when the mark is drawn on a plot. Note: Marks can be spread over several figures (ie several windows in matplotlib).
• View: this is a kind of matplotlib axis, with some enhancement like the ability to draw several layers. All axis functionalities are not available in the view, use a mixin (see pynax.view.colorbar_mixin) to add them easily.