This blog post is aiming to provide a step-by-step guide to perform advanced analytics on swimming data, captured by Apple watch. Microsoft PowerBI and Python on Jupyter Notebook are the primary tools to prepare, analyze and visualize the data.
You will learn how to export the workout data efficiently to your PC, make necessary data transformation, and understand what metrics and dimensions are available. Then I will walk you thru how to analyze the data to answer typical questions related to why certain behaviors happened. You will then see my preliminary attempt to use advanced analytics tools to predict future swimming performance.
Most importantly, you will find quite a few reference articles related to this topic, hopefully fulfilling your intellectual curiosity.
It is also the #3 articles of a series, the previous articles can be found here:
I have been using Microsoft product for over 20 years, starting with MS-DOS, and Windows 3.1. Microsoft PowerBI is my favorite software so far: It is simple, updated frequently, and has the right amount of configuration to “get things done”. The user community used to be small back in 2015, but it is catching up quickly with the forum, free resource (YouTube search “PowerBI”) and paid resource (high-quality material like SQLBI and PowerPivotPro) .
At the same time, as an analytics practitioner, I also understand it is vitally important to keep a mindset of “software independent”. What matters the most is the ability to deliver high-quality deliverable in the most efficient way. A technology tool like PowerBI can help to achieve that, however, we need to be cautious of falling in love with the software, not your own craft and the value delivered.