trainlm predict value & switch

output of trained network is y from

load data

Tb =  readtable('train1.csv','PreserveVariableNames',true)
x = Tb(:,(1:562))%data.simplefitInputs'; 
t = Tb(:,(563:563));%data.simplefitTargets';

[net,tr] = train(net,x,t);
y = net(x);

as target were classes 1,2,3,4,5,6

why output is 1.123 , 4.454, 5.6575 etc not same classes as input?

and how to convert output y to predicted value ?

YPredicted = classify(net.myNet,y) 

returns error:

SWITCH expression must be a scalar or a character vector.

Error in network/subsref (line 173)

switch (subs)

my final output would be confusionmatrix

plotconfusion(YTest,YPredicted)

Answer 1

[Mahesh Taparia]

Hi

In my understanding, you want classification output and you are getting regression output. Convert the variable 't' into a categorical array before training the model and check if it is working or not, i.e

t = categorical(Tb(:,(563:563)));

[Tomasz Kaczmarski]

apologise it will not work

code

%Train= load('TrainSetArray.mat')
%Target = load('TargetSet.mat')
% avaliable under
%https://drive.google.com/open?id=1CzfnIY5DZqcu8Vt-zxM3I6sVLnzLhwK5

x = Train';
t = categorical(table2array(Tb(:,(563:563))));
trainFcn = 'trainlm';  % Levenberg-Marquardt backpropagation.

% Create a Fitting Network
hiddenLayerSize = 1;
net = fitnet(hiddenLayerSize,trainFcn);

% Setup Division of Data for Training, Validation, Testing
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;

% Train the Network
[net,tr] = train(net,x,t);

%plot confusion chart across categories
YPredicted = classify(net.myNet,y) 
plotconfusion(y,YPredicted)

Error in nntraining.setup (line 77)

[net,data,tr,err] = setupPerWorker(net,trainFcn,X,Xi,Ai,T,EW,enableConfigure);

train and target mat files avaliable under below link

https://drive.google.com/open?id=1CzfnIY5DZqcu8Vt-zxM3I6sVLnzLhwK5

[Mahesh Taparia]

Hi

The below is the working code for classification of your application. Hope it will helps.

Train= load('TrainSetArray.mat')
Target = load('TargetSetArray.mat')
% avaliable under
%https://drive.google.com/open?id=1CzfnIY5DZqcu8Vt-zxM3I6sVLnzLhwK5

x = Train.TrainSetArray;
t = Target.TargetSet;
t1=ind2vec(t');
trainFcn = 'trainlm';  % Levenberg-Marquardt backpropagation.

% Create a Fitting Network
hiddenLayerSize = 1;
net =  patternnet(hiddenLayerSize,trainFcn);

% Setup Division of Data for Training, Validation, Testing
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;

% Train the Network
[net,tr] = train(net,x',t1);

y = net(x');%%%%% include the test vectors here
perf = perform(net,t,y);
classes = vec2ind(y);

[Tomasz Kaczmarski]

Hi

apologise but its not working im getting

In test (line 14)

Error using bsxfun

Non-singleton dimensions of the two input arrays must match each other.

Error in gsubtract>calc_general (line 40)

c = bsxfun(@minus,a,b);

additionally im trying to achive is confusion matrix across classes mine code is below

%PREPROCESSING
%load data
Tb =  readtable('train1.csv','PreserveVariableNames',true);
Tbv =  readtable('test1.csv','PreserveVariableNames',true);

%merge both datasets they can be splitted to train target later via dynamic
%parameter
Tbt = union(Tb, Tbv)


%split into train target
TrainSet = Tbt(:,(1:562))%data.simplefitInputs'; 

TargetSet = Tbt(:,(563:563));%data.simplefitTargets';
TargetSet= table2array(TargetSet)

%make some noise
TrainSetArray = table2array(TrainSet)
noiseSignal = cos(5 * pi * 100 * TrainSetArray)+sqrt(5) * randn(size(TrainSetArray));
%TRAINING
x = TrainSetArray';
%x = noiseSignal';
t = TargetSet';
%t = categorical(Tb(:,(563:563)));FAILING DURING TRAINING

trainFcn = 'trainlm';  % Levenberg-Marquardt backpropagation.

% Create a Fitting Network
hiddenLayerSize = 10;
net = fitnet(hiddenLayerSize,trainFcn);

% Create a Fitting Network
hiddenLayerSize = 1;
net = fitnet(hiddenLayerSize,trainFcn);

% Setup Division of Data for Training, Validation, Testing
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;

% Train the Network
[net,tr] = train(net,x,t);

%get output
y = net(x);


%get test set
testX = x(:,tr.testInd);
testT = t(:,tr.testInd);
testY = net(testX);
%PLOTTING CONFUSION MATRIXacorss classes as
% https://uk.mathworks.com/help/deeplearning/ref/plotconfusion.html

YPredicted = classify(net.myNet,testX)%Line Causing error SWITCH expression must be a scalar or a character vector.
plotconfusion(testy,YPredicted)