how to generate simple feed-forward code with accuracy?
显示 更早的评论
Hai. I am new to Matlab and Artificial Intelligent field. Therefore, I need your kind help to generate a simple feed forward for classification. I tried to use simple feed forward coding in Matlab R2012b but I do not know how to check the accuracy. I ve tried different code but it always lead to many errors. I do not know how to fix. I have attached my extracted features data. I need to classify 4 classes.
Previously, I have used this feed-fwd coding but i dont know how to check the accuracy. Plus, what is mean by the performance value?
clear all;
clc;
x = xlsread('mmg_selinp.xlsx');
t = xlsread('mmg_masout.xlsx');
x = x';
t = t';
net = feedforwardnet([15 10],'traingdx')
net = train(net,x,t);
view(net)
y = net(x);
perf = perform(net,y,t)
Sorry, i am really new to Matlab and AI. Your help is highly appreciated. I am not sure whether there s problem with my data. Thank you in advance.
采纳的回答
Greg Heath
2015-4-4
Use patternnet for classification.
help patternnet
doc patternnet
For c classes, the target columns are obtained from eye(c).
Search the NEWSGROUP and ANSWERS using
greg patternnet
Thank you for formally accepting my answer
Greg
PS Include "neural" in you tags
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
