The conference will take place on April 3rd, 2015.

Here is the overall schedule:

8:30 Doors open
9:00 Introduction by the organisers
9:10 Opening Keynote (Gael Varoquaux)
10:00 Track 1: A gentle introduction to PyData
Alexandre Gramfort, then Emmanuelle Gouillart
Track 2: PyData in the real world
Ian Ozsvald
10:45 Coffee break
11:15 Track 1 - continued
Gilles Louppe
Track 2 - continued
Chloe-Agathe Azencott, then Camilla Montonen
12:00 Track 1 - continued
Joris Van Den Bossche
Track 2 - continued
AXA Data Innovation Lab
12:45 Lunch break
14:00 Track 3: High Performance PyData
Niels Zeilemaker
Track 4: Industrial and business case studies
Jean-Paul Smets and Sébastien Robin
14:45 Track 3 - continued
Serge Guelton and Pierrick Brunet
Track 4 - continued
Jean Maynier, then Fabien Mangeant et Vincent Feuillard
15:30 Track 3 - continued
Antoine Pitrou
Track 4 - continued
Julien Sananikone, Benjamin Guinebertière, Samuel Charron, Thomas Cabrol
16:15 Coffee break
16:45 Track 3 - continued
Kirill Smelkov
Track 4 - continued
Clément Jambou
17:30 Closing Keynote (Francesc Alted)
18:30 Closing cocktail
19:30 End of the conference


(See also the detailed program with talks abstracts and speakers bios.)

Keynote speakers

Two keynote speeches by leaders of the PyData community, to get us inspired by the bright future of Python in Big Data and Machine Learning:

  • Gaël Varoquaux (INRIA): “scikit-learn for easy machine learning: the vision, the tools and its development”

  • Francesc Alted (UberResearch GmbH), “New Trends In Storing And Analyzing Large Data Silos With Python”.


16 talks and roundtables that will cover a broad range of topics: high performance computing (HPC) with Python, machine learning, statistics, dealing with messy data, industry use cases…

All talks are 40 minutes long except noted otherwise.

Track 1 (morning): A gentle introduction to PyData technologies

  • 10h-10h45:

    • Alexandre Gramfort (Telecom ParisTech): “Linear predictions with scikit-learn: simple and efficient” (30’)
    • Emmanuelle Gouillart (CNRS): “introduction to scikit-image” (15’)
  • 11h15-12h:

    • Gilles Louppe (CERN): “Tree models with scikit-learn: great learners with little assumptions”
  • 12h-12h45:

    • Joris Van Den Bossche (Ghent University): “Introduction to Pandas”

Track 2 (morning): PyData in the real world

  • 10h-10h45:

    • Ian Ozsvald (Mor Consulting): “Cleaning Confused Collections of Characters”
  • 11h15-12h:

    • Chloe-Agathe Azencott (Mines ParisTech): “Reaching your DREAMs with Python” (15’)
    • Camilla Montonen: “Rush Hour Dynamics: Simulating and visualising commuter flow through the London Underground using graphtool and bokeh” (30’)
  • 12h-12h45:

    • Data scientists from AXA Data Innovation Lab (Axa): “Whitening the blackbox : why and how to explain machine learning predictions”

Track 3 (afternoon): High Performance PyData

  • 14h-14h45

    • Niels Zeilemaker (GoDataDriven): “Embarrassingly parallel database calls with Python”
  • 14h45-15h30

    • Serge Guelton and Pierrick Brunet (QuarksLab): “Pythran: Static Compilation of Parallel Scientific Kernels”
  • 15h30-16h15

    • Antoine Pitrou (Continuum Analytics): “Numba, a JIT compiler for fast numerical code”
  • 16h45-17h30

    • Kirill Smelkov (Nexedi): “Out-of-core NumPy arrays without changing your code with wendelin-core”

Track 4 (afternoon): Industrial and business case studies

  • 14h-14h45

    • Jean-Paul Smets and Sébastien Robin (Nexedi): “Industrial Monitoring with the Wendelin Big Data platform”
  • 14h45-15h30

    • Jean Maynier (Kpler): “Python, SQLalchemy and Scrapy for real-time data processing at Kpler” (20’)
    • Fabien Mangeant et Vincent Feuillard (Airbus): “scikit-learn for predictive maintenance at Airbus” (20’)
  • 15h30-16h15

    • Julien Sananikone (PriceMinister), Benjamin Guinebertière (Microsoft), Samuel Charron (Data Publica), Thomas Cabrol (Dataiku): “Industrial uses of scikit-learn” (business roundtable)
  • 16h45-17h30

    • Clément Jambou (Lyft): “Using Python and Data science to tackle real-time transportation problems at Lyft””