Abstract

The Python programming language offers various features that make it an invaluable part of the scientific programmer’s toolbox. These features include the expressiveness of a modern object oriented language, a large library of functions for statistics, signal processing, numerical optimisation, linear algebra, etc., and active developer and user communities.

For the uninitiated, the talk will start with a brief introduction to the Python programming language and the SciPy library of functions. Thereafter I will present highlights of the recent Advanced Scientific Programming in Python course held in Trento, Italy. A wide range of topics will be covered—including multidimensional arrays, working with massive datasets, source and version control, parallel processing (in shared memory, computing cluster, and GPU flavours), interleaving C++ and Python code, profiling, debugging, and test-based development. 90 minutes is not enough time to do more than scratch the surface of each of these topics, so I will aim to motivate the importance of each of the topics—using examples where possible—and provide references for those who wish to learn more.

Should you wish to do some pre-reading on Python, a friendly introduction with the focus on scientific programming can be found at https://portal.g-node.org/python-autumnschool/_media/pythonscientific.pdf