If you are doing probabilistic programming you are dealing with all kinds of different distributions. That means choosing an ensemble of right distributions which describe the underlying real-world process in a suitable way but also choosing the right parameters for prior distributions. At that point I often start visualizing the …

This talk presented at the PyData Amsterdam 2016 explains the idea of Bayesian model selection techniques, especially the Automatic Relevance Determination. The slides of this talk are available on SlideShare.

Even in the era of Big Data there are many real-world problems where the number of input features has about …

Often I get the question as a Data Scientist what the Python Data Science Stack actually is and where a beginner should start to learn. The Python ecosystem, especially around the topics such as data analytics, data mining, data science and machine learning is so vast and rich that it …

In this presentation given at the EuroPython 2015 in Bilbao, I show how the hangups library can be used in order to write a small chatbot that connects to Google Hangouts and reminds you or someone else to take his/her medication. The secure and recommended OAuth2 protocol is used …

Scikit-Learn is a well-known and popular framework for machine learning that is used by Data Scientists all over the world. In this tutorial presented at the EuroPython 2014 in Berlin, I show in a practical way how you can add your own estimator following the interfaces of Scikit-Learn. First a …