Inputs and targets have different numbers of samples.

can you explain me why i could get this error ?

if true
Error using network/train (line 272)
Inputs and targets have different numbers of samples.
Error in ayobisa (line 49)
net = train(net,data_latih,target_latih);
end

this is my code:

if true
  for loop = 1:5
image_folder = 'E:\23052018 Revisi\data sample2\train\t';
filenames = dir(fullfile(image_folder, '*.jpg'));
total_images = numel(filenames);
% temp = zeros(70,35);
data_latih = zeros(total_images,35);
for n = 1:total_images
  jpgFilename = sprintf('%d.jpg', n);
  full_name= fullfile(image_folder,jpgFilename);
  I = imread(full_name);
  %grayscaling
  abu =  rgb2gray(I);    
  %thresholding
  level = graythresh(I);
  im = imfill(I,'holes'); %Inpaint noisy areas
  bw = im2bw(im,level);
  %cropping
  icrop = imgcrop(bw);
  %resizing
  isize = imgresize(icrop);
  [r,c] = size(isize); 
  %image vector
  %plotchar(isize);    
  %Reshaping
  ishape = reshape(isize,[35,1]);
  ishapet = ishape';
  data_latih(n,:) = ishapet;
end
target_latih = zeros(total_images,1);
target_latih(1:17,:) = 1;
target_latih(18:33,:) = 2;
target_latih(34:40,:) = 3;
target_latih(41:75,:) = 4;
net = newff(data_latih,target_latih,[10 5],{'logsig','logsig'},'trainlm');
net.trainParam.epochs = 1000;
net.trainParam.goal = 1e-6;
net = train(net,data_latih,target_latih);
output = round(sim(net,data_latih));
save net.mat net
[m,n] = find(output==target_latih);
akurasi = sum(m)/total_images*100
end
end

Answer 1

[Greg Heath]

In general:

 For I-dimensional input vectors and O-dimensional 
target/output vectors

 size(input)  = [ I N ]
 size(target) = [ O N ]

Hope this helps,

Thank you for formally accepting my answer

Greg

[Annisa Mujahidah]

so can you please tell me what should i do to fix this problem ? im sorry im a matlab newbie ><

[Greg Heath]

1. Find out WHY you have a different number of inputs and targets

2. If additional data is not available remove data so that the numbers are equal.

Hope this helps.

Thank you for formally accepting my answer

Greg

[Walter Roberson]

The most common cause of this problem is that your input needs to be transposed. Each column needs to be a different sample. A lot of data arrives with each row being a different sample.

[Desmond Molloy]

Does not work.

[francisco caldeira]

for me

transpose()

didn't worked.

[Walter Roberson]

What is size() of your input? What is size() of your target? What is class() of your input? What is class() of your target?