Closed Joey77d closed 1 year ago
Dear Bo Sun,
In order to extract the associated weight to each channel you should select the "Weight vector" as an output in the study configuration window (this checkbox is located in the Classification model tab). Once the analysis is computed and completed you can plot the weight vector as follows:
Please let me know if this solves your problem.
Best, David
Dear Bo Sun, you can find a quick tutorial for the feature contribution analysis in the following link: https://github.com/dlopezg/mvpalab/wiki/Feature-Contribution-Analysis
Best, David
Dear David,
Thank you very much for your guidance. My problem has been solved.
Dear Bo Sun,
In order to extract the associated weight to each channel you should select the "Weight vector" as an output in the study configuration window (this checkbox is located in the Classification model tab). Once the analysis is computed and completed you can plot the weight vector as follows:
- Open the "Plot result utility".
- Click on "Weight analysis" tab.
- Select the result file containing the weights vector time series (it should be located on result/time_resolved/wvector/haufe_corrected/result.mat).
- Configure the resulting plot or animation (start and end time, subject to plot, animation or temporal average, etc.)
- Click on "Plot" button.
Please let me know if this solves your problem.
Best, David
Dear David,
Thank you very much for your guidance. My problem has been solved.
Dear David, I really like this MVPALAB toolbox. I want to know how to analyze the weight amplitude of each channel sorted by its importance averaged across participants. I want to do the same analysis as Figure 13 e and f in your paper. I'm sorry, but I can't seem to find this analysis in the tutorial. I hope to get your advice.
Your paper:López-García, D., Peñalver, J. M., Górriz, J. M., & Ruz, M. (2022). MVPAlab: A Machine Learning decoding toolbox for multidimensional electroencephalography data. Computer Methods and Programs in Biomedicine, 214, 106549.
Bo Sun