Author: Robert Nikutta
Version: 20140416
License: BSD 3-clause, please see LICENSE file
Implements NdimInterpolation class and helper functions. Performs fast multi-dimensional interpolations of models (in arbitrary dimensions) on rectilinear grids (i.e. the pre-computed input grid of pivots need not be equidistantly spaced, or regular, but only orthogonal). Especially useful for SED interpolation, since by default the interpolation is for a 1-d input array of values (think of wavelength list). Implemented schemes: multi-linear, and cubic-spline.
A (slightly) interactive example is provided in example.py . Run like this:
ipython --pylab
In[0]: import example
In[1]: ip, datacube, theta, mywave = example.example()
example.png (shown above) is a preview of what you will see.