I’m using your package and sPLS-DA model to classify bacterial and viral species according to group. I’ve followed the example reported in the srbct-case-study for my data and i’ve split my dataset n=124 into training 31 and test 93. I’ve tuned and cross-validate my model but when I run the script for prediting my test dataset i obtain a confusion matrix with lower observations than my training test (71).
Can you help me to interpret/fix this problem?
Thank you for your support,