41st meetup



NOTE: a valid photo ID is required by building security. Please use your full real names when signing up, otherwise you may be refused entry!



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

We are issuing tickets via a lottery - if you want to be in with a chance of a place - sign up for the waitlist! The lottery will be run approx 1 week before the meetup, and we will re-run the lottery to fill any spaces that free up or use the waitlist towards the time of the event.


Sumanas Sarma on word2vec encodings to build a movie recommendation app
Movie recommendation systems usually base their recommendations on movie metadata - the actors, director, etc along with data about the viewers - ratings / likes. In this talk we examine the use of existing techniques in machine learning (TF-IDF, Word2Vec, Bag-of-Words, k-NN) to build a different type of recommendation engine. The talk will consist of a brief introduction to word vectorisation before stepping through a series of demos that show the stages of building a proof-of-concept movie recommendation web-app. Prior knowledge of ML is not required.

James Lumley on Will Python eat Biotech?
A walk through a knowledge system for automatically suggesting 'what to make next' as used by small molecule drug hunting teams. A look at the python ecosystem that is used to automatically generate these point in time ideas. Further python related AI tools transforming drug discovery research and a perspective on whether python driven autonomous AI will entirely invent your next medication…

Lightning talks:

Robin Cole (https://twitter.com/robmarkcole) on Data science in the home with Home-assistant
Home-assistant (https://home-assistant.io) is an open source, python 3, home automation hub, typically running on a raspberry pi in the users home. Home-assistant can connect with with around 900 different types of sensors, switches and services (collectively referred to as entities), allowing event driven automations to perform actions such as switching on the lights when the user gets home and it is dark. The state of the system (entities and events) is continually logged to an SQL database on the hub, allowing the study of the interaction of the user with their home environment. Products such as the Nest Learning Thermostat demonstrate the benefits of applying data science in the home, but this potential is largely untapped and there is significant room for innovation in this space. For example, could you use the data collected in the home to detect when a water pipe has sprung a leak, saving the homeowner hassle and the insurance company money? Could you identify the compounding factors for an allergy, and advise the homeowner on changes to alleviate their suffering? What could a true smart home, powered by your data science insights, be capable of?

Jonathan Street (http://jonathanstreet.com) on Applying Deep Learning to Histology Images
Accurate detection and counting of glomeruli in renal biopsies is important for diagnostic accuracy but errors can be as high as 50%. Despite challenges due to a modest sample size and images in excess of 5 billion pixels, a fully convolutional neural network is showing great promise including identifying glomeruli previously missed during human annotation.


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

Please unRSVP in good time if you realise you can't make it. We're limited by building security on the 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 5th!