About neural network : newbie question
1 Ansicht (letzte 30 Tage)
Ältere Kommentare anzeigen
Parkz Leingodz
am 18 Dez. 2013
Kommentiert: Parkz Leingodz
am 20 Dez. 2013
I am a new to matlab. I just covert my image to matrix and want to train it. here is my coding:
P= [];
for i =1:15
str=strcat('E:Skripsi\crop2\image',int2str(i),'.bmp');
eval('img=imread(str);');
img=reshape(img,[],1);
P= [P img];
end
T=[1 1 1 1 1 1 1 1 1 1 0 0 0 0 0];// if true image i make it 1 , negative image make it 0
p = num2cell(P,1);
t = num2cell(T,1);
net = newff(minmax(p), [3 1], {'logsig' 'purelin'});
net.trainParam.epochs = 10;
net = train(net,p,t);
Y=sim(net,p)
these is what i got when i compile it
[0.9794] [0.9974] [1.2784] [1.0014] [1.0058] [0.9904] [0.7293] [1.0001] [1.0106]
Columns 10 through 15
[1.0002] [0.0618] [3.3131e-04] [9.9559e-04] [0.0134] [0.0178]
then my problem is what should i do to make the result success i train in percentage with error below 1% (what i want: example i train 10 true image to 1 and one result of my train is 0.798 then accurate =90%) thanks for the help.
0 Kommentare
Akzeptierte Antwort
Greg Heath
am 19 Dez. 2013
There is no reason to use cells or only train for 10 epochs.
Use round to convert the output to {0,1}
Thank you for formally accepting my answer
Greg
5 Kommentare
Weitere Antworten (0)
Siehe auch
Kategorien
Mehr zu Image Data Workflows finden Sie in Help Center und File Exchange
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!