Open LINHYYY opened 1 week ago
I've successfully gotten the result files of 'emg_features' through the code: 'Heat.mat', 'emg_features_examples.mat'. I observed that the 'example' in 'emg_features' is the shape of x 9 6. But I don't know what it is made up of and how to use these two MAT files for regression prediction. I'm just an undergraduate student now, it's a bit difficult for me, I'm very much looking forward to receiving replies and help from the author or other friends, thank you.
Haha I'm back, and after three days of research in my spare time, I came up with a more credible answer to 'x*9*6' of 'examples.shape'. The first thing that is obvious is the 'x', which represents the total data volume of the sample data, which is easy to figure out in the code.The second is '9', I looked at the following values in the variable of 'examples' in matlab, combined with the '.csv' file should correspond to the number of nine electrodes in the original data. Finally, there is the '6', which is the shape that puzzles me the most, and after I have looked through the paper many times, I found that under Figure 9, there is an introduction to the 'Du's sEMG feature', which states that the six characteristic variables are: IEMG, VAR, WL, ZC, SSC, WAMP. Just now I found this 2006 paper and determined that this should be what the '6' means.
But I'm not sure if I'm right in my thinking and haven't tried to do regression model training yet. So if there are any mistakes, please feel free to point them out, and I will apply to contribute my code to this project after the reproduction, including some of the code files I wrote for some other issues such as dataset format and so on. Many thanks to the author for his open source and contributions.
Hello, I take the liberty to bother you again, I wonder if it is possible to expose or provide the code for the prediction of the Gaussian process regression model? I didn't find the corresponding content in your exposed code, and I'd love to learn about this part of the implementation. Thanks again.