In January, I bought a Fitbit Charge HR wristband which, besides counting my steps, also monitors my heart rate, tracks my sleep and more. One reason to buy the device was that I wanted to analyze and visualize my own Fitbit data and thus get a better understanding of my day-to-day activities and simply have some fun playing with data I generated myself. I am currently thinking about writing a series of blog posts describing how I progress with my analysis and what I learn along the way.
I created a GitHub repository to share my code and plots which I will continuously update as I progress. The first step of the analyses was to download my Fitbit data in a programmatic way. Although the Fitbit website allows users to easily download their data, writing a script which downloads the data itself is of course much better to maintain in the future. To download my data, I used the python-fitbit Python client which makes it easy to access the Fitbit API using Python. I documented my workflow in a Jupyter notebook using the Python3 kernel. I invite you to have a look if you are interested.
Next steps in this project will involve visualizing and modelling of the data and I will update my blog once I have something to write about. Please comment below or write me on Twitter if you have any question or suggestions regarding my little Fitbit project.
Update 28 March 2016: Analyzing my Fitbit data. Step 2 - Cleaning step and sleep data and looking for trends is now available.