Treffer: Porting code from Matlab to Python

Python is becoming a popular language for scientific applications and is increasingly used for high performance computing. In this course we want to introduce Matlab programmers to the usage of Python. Matlab and Python have a comparable language philosophy and Python offers complementary alternatives for optimizations and parallelization. Python also provides high portability and flexibility (interaction with other open source and proprietary software packages) without any licensing costs. The training course will be divided into three stages: First, attendants will learn how to do a direct translation of language concepts from Matlab to Python. Then, optimization of scripts using more Pythonic data structures and functions will be shown. Finally, code will be taken to the supercomputers where basic parallel programming (MPI) will be used to exploit parallelism in the computation. The course will focus on numerical and statistical analysis as well as on image processing applications. This course involves theoretical and hands on sessions which will be guided by experts in Python, Matlab and High Performance Computing. Attendants are highly encouraged to bring their own Matlab scripts.