我找到了答案(输出和yy相等):
模拟一个神经网络
清晰的
关闭
clc
rng默认的
[x,y] = simpleclass_dataset;
神经元= 10;
网络=前馈网络(神经元);
Net = train(Net,x,y);
输出= sim(net,x);
% outputs = round(outputs);
图中,plotconfusion (y,输出)
w1 = net.IW{1,1};
w2 = net.LW{2,1};
B1 = net.b{1};
B2 = net.b{2};
Xx = x;
为ii = 1:length(net.inputs{1}.processFcns)
xx = feval(net.inputs{1}.processFcns{ii}，.．.
“应用”xx, net.inputs {1} .processSettings {2});
结束
A1 = tanh(w1*xx + b1);
Yy =纯蛋白(w2*a1 + b2);
为mm = 1:长度(net.outputs{1,2}.processFcns)
yy = feval(net.outputs{1,2}.processFcns{mm}，.．.
“反向”yy, net.outputs {1,2} .processSettings {mm});
结束
图,plotconfusion (y, yy)
=)
