建立神经网络输入参数过多怎么办

yuchaowangok

>  net=newff(p',t,[300,25,1],['tansig','tansig','purelim'],'trainrp','mse');
??? Cell contents reference from a non-cell array object.

5 j' |. M' d+ g# nError in ==> newff>new_5p1 at 199
1 b3 e8 e3 [: N  j) w3 ^9 b' x# y# q5 u  if (length(tf) < i) || all(isnan(tf{i}))

" O$ U$ x' N2 G: B3 P$ wError in ==> newff>create_network at 130
) W& B1 U8 K& w/ ~: A  net = new_5p1(varargin{:});
9 W: I* J& E  g4 e  m3 B
Error in ==> newff at 102
  out1 = create_network(varargin{:});

yuchaowangok

我的这个网络输入参数有30个，但输入元素个数是300个

darker50

   一般就是降维，让参数变少呗！