Prev: Unable to locate decompressor Indeo5 to decompress video stream
Next: matlab R2009b + parfor + SGE (Sun Grid Engine)
From: Mano Samuel on 25 Jan 2010 05:04 hola , I tried the nftool to observe the weight values . So I arranged inputs and targets as following, Inputs - 20 x 70 samples representing 70 samples of 20 elements Targets - 1 x 70 representing 70 samples of 1 element each. Number of neurons in hidden layer = 10; When i go on observe the weight values, it reads net.IW = [10x8 double] [] and net.LW = [] [] [1x10 double] [] i was expecting net.IW = 10 x 20 , because net3.inputs{1}.size * length(net3.inputWeights{1,1}.delays) = 20. Not sure why this happens. Hopefully someone can help me out. Thank you
From: Mano Samuel on 25 Jan 2010 23:20
"Mano Samuel" <manosamuel(a)rediffmail.com> wrote in message <hjjqai$4g8$1(a)fred.mathworks.com>... > hola , > > I tried the nftool to observe the weight values . So I arranged inputs and targets as following, > > Inputs - 20 x 70 samples representing 70 samples of 20 elements > Targets - 1 x 70 representing 70 samples of 1 element each. > Number of neurons in hidden layer = 10; > > When i go on observe the weight values, it reads > > net.IW = [10x8 double] > [] > > and > > net.LW = [] [] > [1x10 double] [] > > i was expecting net.IW = 10 x 20 , because > > net3.inputs{1}.size * length(net3.inputWeights{1,1}.delays) = 20. > > Not sure why this happens. Hopefully someone can help me out. Thank you never mind .. answer found ... |