Recently I decided to switch all my laboratory code from Matlab to Python, mainly to get rid of the elevated costs associated to Matlab licence maintenance, and also to learn a new language that I could also use outside my academic work. To date, I haven’t been disappointed.
But to my surprise, I became much excited about this completely free environment. I am completely amazed by the quality of the scientific python environment: python, scipy, numpy, pandas, matplotlib, spyder, jupyter, etc. I also realize that from now on, I am and will be standing on the shoulders of many, many persons who gave away their time and skills to build this huge environment.
This completely changed my relation to my own code, and I decided to progressively release most of my laboratory code in open source. By doing so, I hope to maximize my contribution to research in biomechanics, particularly in my own fields of rehabilitation and adaptive sports that deserve to be studied extensively.
This first 0.1 release of Kinetics Toolkit (ktk) addresses two things that I find were missing for biomechanics study using python:
- A class that is generic, simple and easy to use for managing timeseries of n-dimensional arrays (e.g., series of markers, forces, transformation matrices) and events, which would be an equivalent of Matlab’s timeseries class. This is now implemented as ktk.TimeSeries.
- An interactive 3d visualizer for markers, rigid bodies and segments, that could be used directly inside IPython/Spyder. This is now implemented as ktk.Player.
Now it’s time to try it!