For this challenge the Fuzzy Labs have competed in a virtual race while wearing some prototype hardware on their shoes, affectionately known as the wearable, my foot! device.
This repository contains the data files recorded by the runners participating in the race. Specifically, accelerometer and gyroscope readings. There's also a Jupyter notebook with some basic data exploration steps.
Runners run along a route of their choosing for a fixed duration of time: 120
seconds. The winner is whoever covers the most distance in the allowed time.
To make the data easier to deal with, runners stand still for a short period of time just before they begin running, and again after they stop.
- For each runner, draw a distance over time graph.
- Simulate the race by visualising distance travelled throughout time.
- Who won the race?
There are three data files in the data directory, recorded by:
- Tom
- Misha
- Matt
Each CSV file has six columns:
- time -- time from start of a recording (units: milliseconds)
- aX, aY, aZ -- acceleration measured by the device in X, Y and Z direction respectively (units: g's, where g = 9.8 m/s^2)
- gX, gY, gZ -- rotation measured by the device around X, Y and Z axis respectively (units: degrees/second)
To get you started we have an example Juptyer notebook with some basic data exploration for the run recordings. Jupyter is a popular data science environment built around the idea of a 'notebook', which is a portable collection of code along with documentation. Even though we've used Jupyter, feel free to use other tools and languages for this challenge if you prefer.
Python 3.
Using Python VirtualEnv is optional but recommended.
First create the virtual environment:
python -m venv env
Then to activate:
source env/bin/activate
You can deactivate the environment any time with deactivate
.
To install Python dependencies:
pip install -r requirements.txt
Activate the virtual environment if you haven't already done so by running source env/bin/activate
. Run jupyter notebook
; Jupyter should launch as a server and open in a browser (if not, just navigate to http://localhost:8888/tree
or the URL reported in the terminal when Jupyter started).
With Jupyter running you can navigate to notebooks/data-exploration.ipynb
. You can execute all the code in the notebook at once, or step through it cell-by-cell.