MPBA at PyCon8 in Florence!

We are Python enthusiasts at MPBA, and our colleagues Ernesto, Valerio, Giuseppe and Gabriele have just come back from their successful experience at PyCon 8, the national conference where professionals, researchers and enthusiasts of this beautiful programming language gather together.

Their commitment has been both as organizers and as speakers:

  • Ernesto Arbitrio: Speakers’ Manager and Speaker

His talk, Indexing and search tons of data with ElasticSearch and Django, explains how to index huge amounts of data, the side effects to avoid when indexing data in search and analytics engine and how to monitor the ELK cluster. Furthermore, the interaction of a Django based web interface with elasticsearch is illustrated.

  • Valerio Maggio: PyData Track Organiser & Speaker

Valerio’s Unconventional Introduction to Deep Learning explains what Deep Learning means in terms of the computational model it requires to be properly effective, pointing out what are the libraries and tools they may leverage for their experiments on GPUs.

Have a look at Valerio’s tutorial Deep Learning, the Keras way: learn how simple and Pythonic is doing Deep Learning with Keras and understand how easy it is to do basic and advanced Deep Learning models in Keras. The tutorial includes an intro to Keras’ main features and an introductory overview of Tensorflow.

  • Giuseppe Jurman: Speaker

Differential network analysis and graph classification: a glocal approach is Giuseppe’s talk about a novel solution in differential network analysis to perform both the network comparison and the classification tasks by introducing the glocal HIM metric for comparing graphs and a graph kernel induced by the HIM measure. We conclude introducing the Open Source implementation of the HIM metrics provided in the R package nettols and in its web interface ReNette

Following the guidelines of Edward Tufte, a master of data visualization, Giuseppe explains in his talk Basic principles of scientific data visualization why the rules of scientific data visualization are important and what happens if some of them are not properly followed and implemented.

  • Gabriele Franch: Speaker

Gabriele gave a talk on Deep Learning for Precipitation and Lightning Nowcasting, describing a deep learning framework applied to weather echo radar and lightning data at regional scale in Trentino-Sudtirol, in the Italian Alps.