Recap of the 15th Data & AI Meetup: Reinforcement Learning; TensorFlow on Azure; Visual Analytics

200 attendees at the 15th Data & AI Meetup at DB Systel in Frankfurt, Germany
200 attendees at the 15th Data & AI Meetup at DB Systel in Frankfurt, Germany

Yesterday we had an amazing Data & AI Meetup in Frankfurt! Let’s have a quick recap!

The venue: DB Systel’s Silberturm

DB Systel kindly hosted the 15th iteration of our Data & AI Meetup on the 30th floor of the Silberturm in Frankfurt, Germany.

Welcome & Intro

Darren Cooper and I had the pleasure to welcome 200 Data & AI enthusiasts! Furthermore, we were happy to announce that our Data & AI Meetup group has 1,070 members and our brand new Data & AI LinkedIn group already has 580 members.

Reinforcement Learning of Train Dispatching at Deutsche Bahn

Dr. Tobias Keller, Data Scientist at DB Systel, showed in his session how Deutsche Bahn aims at increasing the speed of the suburban railway system in Stuttgart (S-Bahn) using Artificial Intelligence. In particular, a simulation-based reinforcement learning approach provides promising first results.

TensorFlow & Co as a Service

Sascha Dittmann, Cloud Solution Architect for Advanced Analytics & AI at Microsoft, showed in his presentation, how TensorFlow and other ML frameworks can be used better in a team through appropriate Microsoft Cloud services. He presented different ways of how data science experiments can be documented and shared in a team. He also covered topics such as versioning of the ML models, as well as the operationalization of the models in production.

Visual Analytics: from messy data to insightful visualization

Daniel Weikert, Expert Consultant at SIEGER Consulting, showed in his session the ease of use of Microsoft Power BI Desktop. He briefly highlighted the AI Capabilities which Power BI provides and showed a way on how to get started with messy data, doing data cleaning and visualize results in an appealing way to your audience.

Speaking at an upcoming Data & AI meetup?

If you’ve dreamed of sharing your Data & AI story with many like-minded Data & AI enthusiasts, please submit your session proposal or reply to the recap tweet:

15th Data & AI Meetup: Reinforcement Learning; TensorFlow on Azure; Visual Analytics

We’d like to invite you to our 15th Data & AI Meetup, hosted at Skydeck @ DB Systel in Frankfurt, Germany.

Agenda:

5:30pm: Doors open

6:00pm: Welcome & Intro
by Alexander Loth, Digital Strategist at Microsoft
and Darren Cooper, Principal Consultant at DB Systel

6:20pm: 🚄 Reinforcement Learning of Train Dispatching at Deutsche Bahn
by Dr. Tobias Keller, Data Scientist at DB Systel

7:00pm: 🚀 TensorFlow & Co as a Service
by Sascha Dittmann, Cloud Solution Architect for Advanced Analytics & AI at Microsoft

7:40pm: 📊 Visual Analytics: from messy data to insightful visualization
by Daniel Weikert, Expert Consultant at SIEGER Consulting

8:30pm: Networking & drinks

9:30pm: Event concludes

DB Systel Skydeck in Frankfurt (previous meetup)
DB Systel Skydeck in Frankfurt (previous meetup)

Sign up on Meetup and join us on Twitter @DataAIHub and LinkedIn!

Do you want to speak at our events? Submit your proposal here: https://aka.ms/speakAI

Machine Learning kompakt: Alles, was Sie wissen müssen

Machine Learning Kompakt Cover und Deep-Learning-Kapitel
Machine Learning kompakt und Blick in das Kapitel “Neuronale Netze und Deep Learning”

Nachdem ich bereits Erfahrung als Buchautor (hier und hier) gesammelt habe, hatte ich kürzlich die Gelegenheit als Technical Reviewer ein sehr spannendes Buchprojekt zu unterstützen. Das Buch Machine Learning kompakt: Alles, was Sie wissen müssen, geschrieben von Andriy Burkov, fand ich dabei dermaßen interessant, dass ich es gerne im Folgenden kurz vorstellen werde:

Machine Learning kompakt von Andriy Burkov ist ein hervorragend geschriebenes Buch und ein Muss für jeden, der sich für Machine Learning interessiert.

Andriy Burkov gelang ein ausgewogenes Verhältnis zwischen der Mathematik, intuitiven Darstellungen und verständlichen Erklärungen zu finden. Dieses Buch wird Neulingen auf dem Gebiet als gründliche Einführung zu Machine Learning zugutekommen. Darüber hinaus dient das Buch Entwicklern als perfekte Ergänzung zu Code-intensiver Literatur, da hier die zugrunde liegenden Konzepte beleuchtet werden.

Microsoft Azure Machine Learning Studio
Microsoft Azure Machine Learning Studio

Machine Learning kompakt eignet sich außerdem als Lehrbuch für einen allgemeinen Kurs zu Machine Learning. Ich wünschte, ein solches Buch gäbe es, als ich studiert habe!

Protip: viele der im Buch vorgestellten Machine-Learning-Algorithmen können Sie einfach und bequem in Microsoft Azure Machine Learning Studio selbst ausprobieren: https://aka.ms/mlst

Machine Learning kompakt: Alles, was Sie wissen müssen (mitp Professional)
  • Andriy Burkov
  • Publisher: mitp
  • Edition no. 2019 (30.06.2019)
  • Broschiert: 200 pages

How to research LinkedIn profiles in Tableau with Python and Azure Cognitive Services

Tableau is using Python to access the Web Services API provided by Microsoft Azure Cognitive Services
Tableau is using Python to access the Web Services API provided by Microsoft Azure Cognitive Services

A few weeks after the fantastic Tableau Conference in New Orleans, I received an email from a data scientist who attended my TC18 social media session, and who is using Azure+Tableau. She had a quite interesting question:

How can a Tableau dashboard that displays contacts (name & company) automatically lookup LinkedIn profile URLs?

Of course, researching LinkedIn profiles for a huge list of people is a very repetitive task. So let’s find a solution to improve this workflow…

1. Python and TabPy

We use Python to build API requests, communicate with Azure Cognitive Services and to verify the returned search results. In order to use Python within Tableau, we need to setup TabPy. If you haven’t done this yet: checkout my TabPy tutorial.

2. Microsoft Azure Cognitive Services

One of many APIs provided by Azure Cognitive Services is the Web Search API. We use this API to search for name + company + “linkedin”. The first three results are then validated by our Python script. One of the results should contain the corresponding LinkedIn profile.

3. Calculated Field in Tableau

Let’s wrap our Python script together and create a Calculated Field in Tableau:

4. Tableau dashboard with URL action

Adding a URL action with our new Calculated Field will do the trick. Now you can click on the LinkedIn icon and a new browser tab (or the LinkedIn app if installed) opens.

LinkedIn demo on Tableau Public

Is this useful for you? Feel free to download the Tableau workbook (don’t forget to add your API key), leave a comment and share this tweet:

#TC18 Wrap-up: Azure SQL Data Warehouse speeds up your Analysis

Benchmark: Microsoft Azure SQL Data Warehouse outperforms Amazon Redshift in TCP-H 30TB
Benchmark: Microsoft Azure SQL Data Warehouse outperforms Amazon Redshift in TCP-H 30TB

Slowly the dust settles after the impressive TC18. During my wrap-up, I remembered the data warehouse benchmarks of the Azure & Tableau session by James Rowland-Jones. Especially because my customers ask me about such performance metrics over and over again.

The first benchmark (graph above) shows how Microsoft Azure SQL Data Warehouse (aka. SQL DW) outperforms Amazon Redshift – in terms of performance and price. While the second benchmark shows further performance tests for Amazon Redshift, Snowflake, Azure, Presto, and Google Big Query:

Benchmark: Microsoft Azure SQL Data Warehouse Gen 2 vs. Amazon Redshift, Snowflake, Presto, Google Big Query
Benchmark: Microsoft Azure SQL Data Warehouse Gen 2 vs. Amazon Redshift, Snowflake, Presto, Google Big Query

Since James’ session is already available on Tableau’s Youtube channel, feel free to watch the entire Azure & Tableau session: