I used this code :
net=feedforwardnet(1);
x=1:10;t=x;
[net,tr] = train(net,x,t);
% the following two values should be equal which is not the case ?!
net(10)
net.LW{2,1}*tansig(net.IW{1,1}*x(10)+net.b{1})+net.b{2}
And when I apply net(10) and I compare it to the explicit value of NN output in terms of biases and transfert function, the resulting values are quiet different.
Please, could you help me clarify why these 2 values are not the same.
Thanks in advance.

 採用された回答

Greg Heath
Greg Heath 2013 年 7 月 10 日

0 投票

Type
net % Without a semicolon
See the defaults that you have not taken into account (e.g., mapminmax)
Thank you for formally accepting my answer
Greg

その他の回答 (0 件)

カテゴリ

ヘルプ センター および File Exchange で Deep Learning Toolbox についてさらに検索

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by