Program sftwolayer | Two layer NN training
[SYNOPSIS]
sftwolayer < dat.rsf label=lbl.rsf valdata=valdat.rsf vallabel=vallbl.rsf weight1=wt1.rsf weight2=wt2.rsf bias1=bs1.rsf bias2=bs2.rsf > loss.rsf weight1out=wt1out.rsf weight2out=wt2out.rsf bias1out=bs1out.rsf bias2out=bs2out.rsf valloss=valloss.rsf lr= niter= act= opt= seed= stop= lossfunc= reg= alpha=
[PARAMETERS]
float   | act | = |   | 	Activation function - 0:sigmoid 1:tanh 2:relu 3:identity
float   | alpha | = |   | 	Regularization coeff. If not, set alpha=0
file    | bias1 | = |   | 	auxiliary input file name
file    | bias1out | = |   | 	auxiliary output file name
file    | bias2 | = |   | 	auxiliary input file name
file    | bias2out | = |   | 	auxiliary output file name
file    | label | = |   | 	auxiliary input file name
float   | lossfunc | = |   | 	Loss function - 0:MSE 1:L1
float   | lr | = |   | 	
float   | niter | = |   | 	
float   | opt | = |   | 	Optimization method - 0:SGD 1:momentum 2:Adam
float   | reg | = |   | 	Regularization - 0:L2 1:L1
float   | seed | = |   | 	
float   | stop | = |   | 	
file    | valdata | = |   | 	auxiliary input file name
file    | vallabel | = |   | 	auxiliary input file name
file    | valloss | = |   | 	auxiliary output file name
file    | weight1 | = |   | 	auxiliary input file name
file    | weight1out | = |   | 	auxiliary output file name
file    | weight2 | = |   | 	auxiliary input file name
file    | weight2out | = |   | 	auxiliary output file name
[DIRECTORY]
user/fomels
