Munich Datageeks - July Edition

It is time for our next Meetup. In July, Munich Network will host us for the second time.

• Please note that the start time was updated to 19:00

• As space is limited this time, please only RSVP if you are sure to make it and please free your spot when you cannot make. Thanks!

Format:

• 2 presentations (each ca. 30-40 min incl. discussion)

• Of course time for networking + food + drinks before, in between and especially after the presentations

• Talks are held in English

The line up:

Jan Stępień - Combating spam, or how I befriended the Killer Rabbit of Caerbannog

Abstract:

You might have received an unwanted email at some point. We all have. According to some studies, between 80 and 90 percent of all email is spam. Those of us with accounts at established email providers can rely on their hosts' filters to keep their inboxes manageable. What if you're hosting your email on your own, though? Off-the-shelf open source solutions are there when you need them, but that's not where the fun is. Combining existing tools, building your own classifiers, and seeing them work in practice is far more exciting. Let me tell you a story, one with rabbits.

Bio:

Jan is a senior consultant at innoQ, where he works with people and with computers. He's based in Munich but you can run into him in Berlin too. When he's not caressing rabbits or filtering spam he can be found co-organising the Munich Clojure Meetup. Pop in, we're a really friendly
bunch.
https://janstepien.com ( https://janstepien.com/ ) and @janstepien on Twitter.

Stefan Coors - shinyMlr

Abstract:

Conducting machine learning tasks can be very tedious. For example the R software repository CRAN serves hundreds of different packages not only for learning algorithms, but also for data preprocessing, hyperparameter tuning and more, each with their own API. Software like caret and mlr in R and scikit-learn in Python collect techniques and methods inside a single package with a unified interface. This makes coding easier, shorter and therefore less error-prone. However, a programming background is still required, of course. To make the wide palette of features available to people with little, or without programming experience, that do not want to pay for enterprise ready machine learning software, we took the idea of mlr one step further by developing an open source graphical user interface, built with the R package shiny, which wraps mlr's functionalities in a web app. With our app you can perform all basic machine learning tasks just with a couple of mouse clicks: From data import and preprocessing, to benchmark experiments, tuning and rendering an interactive report. In this talk we will present our app step by step, illustrate the workflow with example data sets and compare our project with similar already existing applications.

Bio: