1.  run GenerateData.sh with force_scale = 2.0


2.  For point-to-point learning
    run NNLearn.sh  
    it has "julia NNLearn.jl 2 spd_Chol_Orth 0"  The source file is in NNLearn.jl, and the number 2 (idx) determines nn layer number in nnutil

3.  For indirect data learning
    a) run PreLSfit.jl
    The source file is NNPreLSfit.jl
    b) run NNTrain.jl
    The source file is Train_NNPlatePull.jl, you can choose the restart output (from the PreLSfit), the current restart_id=5 (5000 iterations)


The present setup is nntype=stiffmat in all files, you can change it to ae_scaled. 