XLVII Evento Porto.Data – 12 Dezembro 2018 | UPTEC

Realiza-se no dia 12 de Dezembro de 2018, o XLVII Encontro da comunidade Porto.Data, que tem como objetivo a partilha de experiências e conhecimento entre os profissionais e entusiastas da plataforma de dados da Microsoft no Porto e desta vez faremos em conjunto com a comunidade de Python. O evento vai ser realizado no auditório da UPTEC, pelas 19:00, sendo a entrada livre para todos os inscritos. Registo gratuito.

Agenda
18:50 – Abertura e recepção.
19:00 – "Machine Learning with Python in SQL Server Ecosystem"
       Diamantino Falcão - Data professional 
19:25 – Coffee break / Q.A / Community News.
19:40 - "Cohort analytics with Redis and bitmapist"
       Roman Imankulov - Web/Python Developer
20:05 – Fim do evento.
20:15 – Jantar livre.

Diamantino Falcão é DBA/BI Developer numa empresa da área financeira 3CTechLab. Licenciado em engenharia informática e com mais de 10 anos de experiência em IT, um entusiasta por plataformas de dados sendo o seu produto de eleição o SQL Server.

This session will walk you through the process of moving data between SQL Server and Python: requirements, data structures, inputs, and outputs. It also explains how to wrap Python code in a stored procedure to build, train, and use machine learning models in SQL Server.

 

 

Roman is the head of the web development team at Doist and one of the organizers of the Python Porto community. He lives in Porto for almost four years, and works with Python for more than a decade. Believes that simple is better than complex and practicality beats purity.

When it comes to data storing and analysis, we get used to think big. Cloud solutions with a monthly price tag of four or five digits are not uncommon on the market. But if you are a startup, you don’t have resources or expertise to maintain such systems, and frankly rarely you need them. This talks is about bitmapist, a Redis-based system for event tracking and cohort analytics. It was created six year ago within a couple of days by one person. Six years from now, bitmapist still occupies one mid-size AWS machine. Every day it grows to a couple of millions of events and provides ad-hoc cohort analytics for any query within 1-2 seconds.