On Tue, May 1, 2007 2:47 pm, [EMAIL PROTECTED] said: > Hi, > > I am considering using Pyro to implement a recurrent fuzzy neural net. I > see > that a recurrent neural net is given and a fuzzy logic set is given. Would > I > then just modify the various weight functions to work with fuzzy logic?
Kwame, That sounds reasonable. Note that Pyro's Conx neural network code only implements Simple Recurrent Networks, but output node activations will be between -1 and 1. -Doug > Sincerely, > > Kwame P.R. > _______________________________________________ > Pyro-users mailing list > [email protected] > http://emergent.brynmawr.edu/mailman/listinfo/pyro-users > -- Douglas S. Blank Associate Professor, Bryn Mawr College http://cs.brynmawr.edu/~dblank/ Office: 610 526 6501 _______________________________________________ Pyro-users mailing list [email protected] http://emergent.brynmawr.edu/mailman/listinfo/pyro-users
