Steps
March 2023
Step Counter Tutorial
This tutorial provides a step-by-step guide for implementing a step count algorithm using Python. It outlines the implementation of a hybrid step count model developed by Small et al. (2023)[1]. It is part of the RMLHDS (Reproducible Machine Learning in Health Data Science) project by OxWearables.
The tutorial is available as a Jupyter notebook on GitHub, and can be accessed via the following link: https://github.com/OxWearables/RMLHDS/blob/main/8_stepcount/Tutorial.ipynb
Overview
This tutorial covers the following topics:
- Downloading of OxWalk dataset
- Data pre-processing
- Peak detection
- Feature extraction
- Walking detection
- End to end model training
- Model evaluation
Requirements
All requirements for starting this tutorial have been prepared in previous sessions.
Usage
To use the step count algorithm, follow these steps:
- Navigate to the 8_stepcount directory:
cd RMLHDS/8_stepcount - Open the Jupyter notebook:
jupyter notebook Tutorial.ipynb - Follow the instructions in the notebook to implement the step count algorithm
Conclusion
This tutorial provides a comprehensive guide for implementing a step count algorithm using Python. By following the steps outlined in the tutorial, you can develop a step count algorithm that can be used to analyze data from wearable devices and provide insights into physical activity levels.