I have released a tutorial of the Spark machine learning library (MLlib). This tutorial is not intended to explain any ML theory, although some theory can be found. The tutorial is more a collection of examples around how to manipulate data structures to feed the algorithms implemented by this library.
The corresponding Jupyter Notebook is available here. If you want to fork the repo:
git clone https://github.com/juanmanuel-tirado/pyspark-tutorial
Additionally, you can see the rendered version below.