I have 1000 data sets. Inputs contain 8 variables. and there are 6 responses of which 4 are important to me. What I am wondering is how to get started say if I just consider one of those 4 as my response variable data. Should I split the training and testing data sets as 70/30. Is that good enough. Is there a simple example available in help files? I am doing this for the first time so the question might be totally dumb so please bear with me. Thanks in advance.