Scientific Programme

The program of the school is devoted to project development for astrophysics & astroparticle. The aim of the school is to provide theoretical and hands-on training on Python development (as well as some training in C/C++ coding for Python).

Prerequisites: intermediate level (or more) in Python is required. Python will be the language of reference during the school. 

Preliminary program:

The program will be as follow:

Project management:

  • Git
  • Continuous integration
  • Class, module & package in Python
  • Debugging & profiling

Python libraries for science:

  • Numpy & Pandas
  • SciPy
  • Matplotlib
  • Astropy

Machine learning

  • Introduction to machine learning theory
  • Machine learning in practice (linear regression, binary classification)
  • From neural networks to convolution neural networks
  • Data science workflow
  • Application to astrophysics and cosmology data
  • Human in the loop: adaptive learning

C/C++ for Python

  • How CPU works (introduction level)
  • Software optimization (vectorization, ...)
  • Wrapping C/C++ code in Python

 

On the last day of the school, we will hear a keynote lecture by Gregory Garcia, from the company Ubisoft (a French video game company) who will present us "A glimpse of game engines":

The presentation will cover goals, features and technical challenges to build such piece of software. It will also give some insights on how we organize development with multidisciplinary, local and remote teams.