Problems with Rprop-Algorithm in batchman-mode

• Messages sorted by: [ date ][ thread ][ subject ][ author ]
• Next message: : "biological simulator ??"
• Previous message: : "Number of hidden units added by Cascade Correlation"
• Next in thread: : "Re: Problems with Rprop-Algorithm in batchman-mode"

Trying to start the training of a neural net by Rprop instead of 
Std_Backpropagation caused the following problem: Although I've 
chosen the default parameters of the manual, there was just no 
training taking place by Rprop (Training took place when Rprop was  
replaced by Std_Backpropagation in the batchfile).

Does anyone have experience how to overcome this problem? - Thank you 
for information.

Kai Hufendiek

---------------------------<< End of Message >>------------------------------

Dipl.-Ing. K. Hufendiek, M.Sc.(UMIST)
Institut fuer Energiewirtschaft und
Rationelle Energieanwendung (IER)
Universitaet Stuttgart
70550 Stuttgart
Germany
Phone: +49-711-78061-58
Fax:   +49-711-7803953

http://www.rus.uni-stuttgart.de/UNIuser/ier/

-----------------------------------------------------------------------------

• Next message: : "biological simulator ??"
• Previous message: : "Number of hidden units added by Cascade Correlation"
• Next in thread: : "Re: Problems with Rprop-Algorithm in batchman-mode"