hi, for the first question :
your training and testing files should be orgnized as follow: [targets, features].
and you should normalize only the features between 0 and 1 but inside the ELM function :
you can download the normalization function from here https://www.mathworks.com/matlabcentral/fileexchange/70175-extreme-learning-machine-with-sub-nets-classify-or-predict?s_tid=prof_contriblnk as "scaledata.m".
then search inside the code were you can found inputs (features) then write (inputs =scale data(inputs,0,1)).
do the same thing for trainig inputs and testing inputs only (do not normalize the targets)
