wcoulton/numpy_f2py_talk

Notes for talk on numpy and f2py

These are notes/example code used for my talk optimizing python code with numpy and f2py. They are mostly meant as a skeleton to give the talk around, not as a replacement for the talk, but I post them here in the hope that they will be useful anyway.

1. numpy_intro.py: Introduction to numpy through many small examples.
2. numpy_performance.py: How fast is numpy compared to python?
3. numpy_f2py.py: Calling fortran from python via numpy. How much is there to gain?
4. fortran.f90: Fortran code used in numpy_f2py.py.

These are mostly self-contained from the talk, and can be run as stand-alone examples, for example python numpy_performance.py. Before running python numpy_f2py.py, the fortran code needs to be compiled. On linux (possibly mac?), this should be as simple as running make.