How do you make for get one/zero output?

the result not is iqual to the vector target because i get 0.5 in place of zero. What function i can add to my code for to get the output iqual to the target?

inputs =

Columns 1 through 11

    6.3200    6.7800    7.5600    6.8400    9.0000    6.7200    6.5200    6.7400    7.0600    8.8400    6.2600
         0         0         0         0         0         0         0         0         0         0         0

Columns 12 through 22

    7.6400    8.1400    7.5200    8.0600    6.4800    6.5000    6.6600    7.2400    7.3200    5.1200    8.2400
         0         0         0         0         0         0         0    1.0000    1.0000    1.0000    1.0000

Columns 23 through 32

    6.7800    8.0200    8.0800    6.6600    7.7800    6.3400    8.3000    9.0000    7.2000    5.7800
    1.0000    1.0000    1.0000    1.0000    1.0000    1.0000    1.0000    1.0000    1.0000    1.0000

target =

Columns 1 through 19

     0     0     0     0     1     0     0     0     0     1     0     0     0     1     0     0     0     0     0

Columns 20 through 32

     1     0     1     0     0     1     1     1     0     1     1     0     1

with the code: net=feedforwardnet(10); net.layers{1}.transferfcn='logsig'; net.layers{2}.transferfcn='logsig'; net.divideParam.trainRatio = 100/100; %net.divideParam.valRatio = 10/100; %net.divideParam.testRatio = 10/100; %net = configure(net,inputs,target); net=init(net); %net.iw{1,1} %net.b{1} %net.lw{2,1} net.trainparam.goal=0.001; net.trainparam.epochs=1000; net.trainFcn = 'trainlm'; %[net,tr]=train(net,inputs,target); net=train(net,inputs,target); output=sim(net,inputs) error=mse(target-output)

I get output: logit1

output =

Columns 1 through 11

    0.5002    0.5002    0.5008    0.5002    0.9855    0.5002    0.5002    0.5002    0.5003    0.9101    0.5002

Columns 12 through 22

    0.5009    0.5027    0.5007    0.5022    0.5002    0.5002    0.5002    0.5062    0.5085    0.5001    0.8629

Columns 23 through 32

    0.5010    0.6980    0.7434    0.5006    0.5679    0.5002    0.9006    1.0000    0.5053    0.5001

error =

    0.2121

Answer 1

[dpb]

disp([i'>7.5 o'>0.55])

?, maybe?

Answer 2

[FRANK]

My dear, I don't understand, could you say me in what place to the code write its?

[dpb]

Any point at which you have the outputs of choice.

It simply demonstrates that you can create a logical vector of 0 and 1 by the comparison of the inputs to a suitably chosen reference level. What you do with that is dependent on your needs.

PS. It would help immeasurably if you would edit your initial question to properly format the code so it's not line-wrapped. Use the {}Code button or just prefix the first line of code with two blanks. Use the preview window to see what it looks like until get it right...

Answer 3

[FRANK]

net=feedforwardnet(10,'trainlm'); net.layers{1}.transferfcn='tansig'; net.layers{2}.transferfcn='tansig'; net.divideParam.trainRatio = 80/100; net.divideParam.valRatio = 15/100; net.divideParam.testRatio = 15/100; %net = configure(net,inputs,target); %net=init(net); net.iw{1,1} net.b{1} net.lw{2,1} net.trainparam.goal=0.001; net.trainparam.epochs=1000; %net.trainFcn = 'trainlm'; %[net,tr]=train(net,inputs,target); net=train(net,inputs,target); output=sim(net,inputs) error=mse(target-output)

[dpb]

Still not formatted, unfortunately...and would be better if edit the original question, then remove this comment (and I'd follow up by cleaning up this one)