home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
HPAVC
/
HPAVC CD-ROM.iso
/
NEUMAP3.ZIP
/
NUMP.ZP
/
MLMP.HLP
< prev
next >
Wrap
Text File
|
1994-09-06
|
2KB
|
69 lines
Outline
1. Multilayer Perceptron (MLP) Options
2. Error Functions
1. Multilayer Perceptron (MLP) Options
a. Train an MLP using backpropagation (BP). Batching, which
denotes the accumulation of weight changes over portions of the
training set before they are used, is an option. The learning
factor changes adaptively.
b. Fast training of MLP networks. Trains networks one or two
orders of magnitude faster than BP.
c. Analyze and prune trained MLPs from BP or fast training.
Produces weight and network structure files for the pruned network,
which can be saved to disk, in the non-demo version.
d. Process data using a trained MLP. Data may or may not include
desired outputs.
e. Create MLP subroutine. Given a network structure file and a
weight file, creates an MLP subroutine in Fortran with a parameter
list that includes only the input array and output array.
f. Create formatted weight file. Given a network structure file and a
weight file, creates a formatted weight file that clearly shows
the different connections and their weights and thresholds.
2. Error Functions
a. The error function that is being minimized during backpropagation
training and fast training is
Nout
MSE = (1/Npat) SUM MSE(k) where
k=1
Npat 2
MSE(k) = SUM [ Tpk - Opk ]
p=1
where Npat is the number of training patterns, Nout is the number
of network output nodes, Tpk is the desired output for the pth
training pattern and the kth output, and Opk is the actual output
for the pth training pattern and the kth output. MSE is printed
for each iteration.
b. Additional errors printed out are defined as follows.
The rms error of the kth output, RMS(k), is SQRT( MSE(k)/Npat ),
where SQRT means square root.
The kth output's Relative RMS Error is
R(k) = SQRT( MSE(k)/E(k) ) where
Npat 2
E(k) = SUM [ Opk-Mk ] and
p=1
Npat
Mk = (1/Npat) SUM Opk
p=1
The kth output's Error Variance is MSE(k)/Npat.