arose13 / gmem Goto Github PK

Hi,

I am wondering if it's possible to use the GMEM algorithm for classification tasks. I wrote some code to classify Parkinson's disease patients and controls in our study, and I was hoping someone could take a look at it.

In our study, we recorded data from 78 individuals (both controls and patients with PD) and performed 12 smooth pursuit experiments. For each person, we performed 12 different experiments, varying in amplitude, speed, and direction of the target. We extracted 70 features for each recorded signal. I have set this data in a data frame (936*73), where the columns are Subject, 70 features, SPT (12 different experiments), and Group ('0' for PD and '1' for control).

I am attaching my code for review. Could someone please take a look at it and let me know if the GMEM algorithm is suitable for this type of classification task?

Thank you very much.
GMEM.zip