29th Meetup

Note: Please use your full real names where signing up, otherwise we have problems with building security.

As always, there'll be free beer and pizza, generously provided by AHL.

---

Main speakers:

Paul Jones (https://twitter.com/paulwilljones) on Functional Python

Although Python is thought of as a procedural and OOP language, there are plenty of features to facilitate the implementation of a functional perspective. Our discussion will encompass the fundamentals of functional programming, whilst demonstrating how we would implement functional techniques in Python.

Deenar Toraskar (https://twitter.com/DeenarToraskar) on 10 things you didn't know you could do with interactive notebooks

Interactive notebooks were born to address the needs of reproducible research in academia. Spark and the big data stack have given a new lease life to interactive notebook interface. The ability to access distributed storage (HDFS) and having access to a Spark cluster means the notebook is very powerful and able to handle the most complex tasks. This makes it a good choice for a variety of tasks you traditionally didn’t associate with an interactive notebook.

Peter Goldsborough (http://www.goldsborough.me/) on A Tour of Tensorflow

A walkthrough of a typical Neural Network implementation with Tensorflow, explaining all the moving parts, concepts and differences to similar libraries as we go along.

---

Lightning Talks:

John Stinson on Rich relational data from thin air: how to fake it

What do you do when you don’t have access to the data you need? Fake it with Python of course! John Stinson will talk about patterns for simulating trends in relational data, based on his recent experience reproducing media-player usage statistics.

Jose Alberto Esquivel (https://github.com/betoesquivel) on Embarrassingly parallel data analytics in Python on > 800 cores with AWS Lambda

If you want to scale up your code to run it on large datasets, don't change it! Keep your beloved DataFrames and Counters. In fact, you can parallelise it on a large number of CPU cores without having to own a large cluster or to rent EC2 nodes. We will show how you can parallelise and run your "big data" experiments in a quick and cost effective way using AWS Lambdas.

Michelangelo Bucci on Graph Knowledge based data science with Grakn.ai

There is no way around it: as the world becomes more and more data rich, the task of managing and wrangling data becomes more and more complex. Knowledge graphs are set to be the next big thing in terms of data integration and search. But what is a knowledge graph and how can you get one? And, more than anything, should you care about it? To try to briefly answer these questions, I will present Grakn.ai, an open-source Knowledge Graph data platform and show the benefits that it can bring to data scientists. These include: easy integration of multiple data sources, a flexible schema to check correctness and express how the information is related, and an intuitive graph query language to extract information across all sources.

---

Logistics:

Doors open at 6.30 (get there early as you have to sign-in via AHL's security), talks start at 7pm, beers from 9pm in the bar. We normally have > 200 folk in the room so there's plenty of people to discuss data science questions with!

Please unRSVP if you realise you can't make it. We're limited by building security on number of attendees, so please free up your place for your fellow community members!

Follow @pydatalondon (https://twitter.com/pydatalondon) for updates and early announcements. See you on the 6th!