sorry, we can't preview this file

...but you can still download BCmodel_RF_20210225_v27.pkl
BCmodel_RF_20210225_v27.pkl (1.3 GB)

Machine learning-based classification model corresponding to the paper 'Machine learning to improve orientation estimation in sports situations challenging for inertial sensor use'

Download (1.3 GB)
software
posted on 13.07.2021, 09:54 by Marit van Dijk
Machine learning model as used in 'Machine learning to improve orientation estimation in sports situations challenging for inertial sensor use'. The model to run the Extended Madgwick filter is based on a random forest algorithm and can be used as explained in Figure 3 in the paper.

Explanation for use:
The model can be loaded and executed in Python using the following code
- RFmodel = pickle.load(open([filename_RFmodel], 'rb'))
- y = RFmodel.predict_proba(X)[:,1]
- if y > 0.5
EFcorrect = 1
else
EFcorrect = 0
X are (normalized) input variables
y are probabilities for EFcorrect (see paper)

Funding

ZonMw project number: 546003002

History

Publisher

4TU.ResearchData

Organizations

TU Delft, Faculty of Mechanical, Maritime and Materials Engineering, department of BioMechanical Engineering Vrije Universiteit Amsterdam The Hague University of Applied Sciences