SciPy 2013 was the twelfth annual Scientific Computing with Python conference, held this year in Austin, TX, from June 24 to June 29. I was unable to attend, so I am giving this report from a distance based on the meeting web page, conversations with attendees and the Twitter feed (hashtag #scipy2013). I was not able to unearth many blog postings – am sure there must have been plenty, and if readers know of any, please send me links to them.
The Conference has special mini-sessions on the application of Python to specific disciplines: Meteorology, climatology, and atmospheric and oceanic science, Medical imaging, Bio-informatics, GIS – Geospatial Data Analysis, and Astronomy and astrophysics. The astronomy talks were as follows:
- Python and the SKA (Simon Ratcliffe SKA South Africa, Ludwig Schwardt SKA South Africa).
- Combining C++ and Python in the LSST Software Stack (Jim Bosch, Princeton University).
- Ginga: an open-source astronomical image viewer and toolkit (Jeschke, Eric, Subaru Telescope, National Astronomical Observatory of Japan)
- Astropy, growing a community-based software system for astronomy (Droettboom, Michael, STScI; Robitaille, Thomas, Max Planck Institute; Tollerud, Erik, Yale University)
- SunPy – Python for Solar Physicists (Mumford, Stuart, University of Sheffield / SunPy)
- Accessing the Virtual Observatory from Python (Plante, Raymond, NCSA/UofIL; Fitzpatrick, Mike, NOAO; Graham, Matthew, Caltech; Tody, Doug, NRAO).
The conference also included specialized tracks on Machine Learning and Reproducible Science, and three keynote presentations:
- IPython: from the shell to a book with a single tool; the method behind the madness, by Fernanado Perez.
- The New Scientific Publishers, by Fernando Perez, and
- Trends in Machine Learning and the SciPy community, by Olivier Grisel.
The program contained an extensive program of tutorials, with presentations on such topics as Statistical Data Analysis in Python, Data Processing with Python, Diving into NumPy code, and Version Control and Unit Testing for Scientific Software (I was delighted to see a tutorial on best practices). Links to videos of these sessions, along with tutorial descriptions an links to documentation, are available from the tutorials page. Here is part 1 of 3 of the Data Processing with Python session, which I particularly liked:
I was hoping that the conference web page would include links to talks, but fortunately many links have been posted on Twitter (check out hashtag #scipy2013 to see the latest). I learned that all videos have now been posted on-line at ow.ly/mBgUl . Here is one talk I particularly enjoyed: Astropy, growing a community-based software system for astronomy, by Erik Tollerud, Michael Droettboom and Thomas Robitaille
I also enjoyed Chris Beaumont’s talk on Multidimensional Data Exploration with Glue: