Academic
Blog
Analytics

Two years data from Withings smart watch

2 mins read time ❤️ Views


The withings smart watch looks like a regular watch. It’s very difficult to tell that it has bluetooth and movement sensors inside. In addition it has long duration battery and it’s very elegant.

I have been using the watch during two years. Here you can find some of the insights that the data that the watch collected tell me.

Calendar heatmap of steps

This calendar plot shows the number of daily steps. Each box is a day. It’s interactive, so you can hover your mouse to read the data. The colour meaning is the following:

  • The white means that there are no steps that have been recorded that day.
  • Light blue means that I didn’t walk a lot that day.
  • Dark blue means that I have been walking a lot that day.

I have been wearing the device very often, so you can notice that there are a lot of blue squares. It’s also interesting to see holidays where I walk more than regular.


Daily steps

This chart shows the number of daily steps in a more regular way. You might see the following:

  • I have been walking more at the end of 2016, at the end of my PhD.
  • My walking record is 37k steps, on 14 august 2016. It was on holidays at Italy.

Steps per day

This chart shows the number of daily steps grouped by weekday. You might notice that Saturday is the day when I walk the most and Monday is a very lazy day.

Steps per month

This chart shows the number of daily steps grouped by month. You might notice some seasonal effects.

  • The summer is very active for me. Being July the month when I walk the most.
  • I tend to be more lazy on winter. November is a month where I didn’t walk a lot in two years.

Conclusions

This data helps me to know myself more. Mondays are not my days. Saturdays and Summer are the periods of the time when I walk more.

The code to process data and plot the chatjs graphs can be found in this python notebook.

In relation with 🏷️ run, python, chartjs, withings:

Marathon training program analysis using python and chartjs

A marathon needs to be prepared for at least 3 months. This post shows my preparation training and a prediction of my time for the Metz Marathon 2017

Running statistics from Santorini experience

In this post I use the javascript library Chartjs to illustrate interactively the statistics of a running activity at Santorini experience 2016

Intelligent Video Analytics using SSD mobilenet on NVIDIA's Jetson Nano

This project implements a deep learning model on a Jetson Nano to count and track people passing in front of a video camera.

Application revenue prediction

This article presents a study on predicting user-generated revenue within a digital application. Using gradient boosting models and a dataset containing user installation features and engagement metrics, the study aims to forecast revenue at day 120. The model's performance and key influencing factors are analyzed, providing insights for optimizing monetization strategies in the digital applications industry.

📅 Oct 7, 2024   kaggle   python
Subscribe to my newsletter 📰

and share it with your friends: