I want to make a presentation with comparison of pandas and pyspark. For this purpose I need some data and ideally a research task that will allow me to:
- Show how joins works in both libraries. So at least the data should consist of two datasets. Ideally data should allow and the task should require some joins/group bys to be performed to show all the spark sql features.
- Compare the speed of the libs.
- Ideally data should come with a research task/tutorial, so I could show PySpark mllib in action
Could someone please suggest a tutorial/kaggle page/anything that will fit my requirements?