jozefcoldenhoff / non-linear-motor-decoding Goto Github PK

Github repository of the thesis: A comparison of non-linear methods for the decoding of motor performance.

• Install Anaconda
• Fork the repository
• Use conda env create -f non-linear-motor-decoding.yml to create a new environment with the required packages
• Use conda activate non-linear-motor-decoding to activate the new environment
• Finally use Jupyter Notebook to open the notebooks

The notebook Binned data creation and analysis.ipynb contains the code used to extract the features from the copy-draw data. It also contains the plots used in the data analysis, as well as the result of the bootstrap test for the difference in medians.

The notebook Method evaluation.ipynb contains the results of the methods on the three different datasets.

The notebook Method evaluation decorrelated.ipynb contains the results of the methods on the dataset with the correlated features removed.