Integrates from a to b, given a array of data points, by using piecewise linear interpolation. Very fast.

You know numpy.interp(x, xp, yp)? It interpolates x using the arrays xp (coordinates, increasing) and yp (values). It does so via a binary search to find the closest coordinate to x in xp.

If you have a large array of x values this can be slow because numpy.interp does a binary search every time. This tiny library assumes that the x values are ordered, and thus continues its search in the vicinity of the last lookup.

This makes this interpolation very fast.

This library (just 1 C function) actually does not just interpolate, but integrates bins, which are defined by a lower bin border and an upper bin border. The integration is linear piecewise.

$ make

And point LD_LIBRARY_PATH to the directory containing it.

See monointerp.py for calling from Python