Use Classification Neural Network Model for another Dataset

1 view (last 30 days)
Dear All; I have built an AI model to classify the data using a dataset. Then i try to test this model to classify an external data set but it does not work properly because the code is not properly made , the code is below , could you help me in that :
clc; clear; close all; %Read The data [x1,TXT,RAW]=xlsread('ALL2.xlsx','lnRe'); [t1,TXT2,RAW2]=xlsread('ALL2.xlsx','OUT2');
x=x1';
t=t1';
% Build the model
net= patternnet ([100]);
% net.divideParam.trainRatio = 70/100;
% net.divideParam.valRatio = 15/100;
% net.divideParam.testRatio = 15/100;
% view(net)
net=init(net);
[net,tr] = train(net,x,t);
% Test the Network [x2,TXT3,RAW3]=xlsread('expsettest.xlsx','Ln(Re)'); [t2,TXT4,RAW4]=xlsread('expsettest.xlsx','out-test'); xt=x2'; tt=t2'; outputs = net( xt); errors = gsubtract(tt,outputs); performance = perform(net,tt,outputs)
figure, plotconfusion(tt,outputs)

Accepted Answer

Greg Heath
Greg Heath on 20 Sep 2015
100 hidden nodes appears to be a ridiculous number.
Why don't you start by just using all defaults.
help patternnet
doc patternnet
Then Search NEWSGROUP and ANSWERS using
greg patternnet
Hope this helps.
Thank you for formally accepting my answer
Greg
  2 Comments
mustafa alnasser
mustafa alnasser on 20 Sep 2015
Dear Greg
Thank you for your help
Let me explain for you my issue:
My problem is in the second part of the code, which is test part . in the confusion matrix , i got unreasonable results in terms of data points for each group , it gave more data that it should be for each group , i feel that i am not programing the test code in right way
Best Regards

Sign in to comment.

More Answers (0)

Categories

Find more on Sequence and Numeric Feature Data Workflows in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!