I am getting an error on the last line saying Input and target has different samples, Any suggestions?

load pumpFeatures.mat

t = table2array(T(:,1:84));

Y = table2array(T(:,85));

% features (X) are stored in the first 84 columns and labels (Y) in the last column of the dataset 'data'

% Split features (X) and labels (Y)

X = table2array(T(:, 1:84));

%Y = table2array(T(:, 85));

% Define the size of the training set (e.g., 70%)

training_ratio = 0.7;

% Generate indices for the training and test sets

cv = cvpartition(size(T, 1), 'HoldOut', 1 - training_ratio);

% Split the features and labels into training and test sets

X_train = X(cv.training,:);

Y_train = Y(cv.training,:);

X_test = X(cv.test,:);

Y_test = Y(cv.test,:);

hiddenLayerSizes = [10];

net = feedforwardnet(hiddenLayerSizes);

net = configure(net, X_train, Y_train);

net.layers{end}.transferFcn = 'softmax'; % Softmax for multi-class classification

% Train the network

net = train(net,X_train,Y_train); --- Error Line Input and Target has different numbetr of samples

Answer 1

[Manikanta Aditya]

Hey Joseph,

The error message you’re seeing typically occurs when the dimensions of the input data and target data do not match. In your case, it seems like the dimensions of 'X_train' and 'Y_train' might not be aligned.

The train function in MATLAB expects the input and target data to have the same number of columns, where each column represents a single sample. However, in your code, 'X_train' is a matrix where each row represents a sample, and 'Y_train' is a vector.

To resolve this issue, you can transpose 'X_train' and 'Y_train' before passing them to the train function. Here’s how you can do it:

% Train the network
net = train(net, X_train.', Y_train.');

This will ensure that both 'X_train' and 'Y_train' have the same number of columns, each representing a single sample. Please give this a try.

Check the following MATLAB Answers which talk about the same error:

• https://www.mathworks.com/matlabcentral/answers/429762-input-and-target-have-different-number-of-sampel
• https://www.mathworks.com/matlabcentral/answers/418894-inputs-and-targets-have-different-numbers-of-samples

Thanks!

[Joseph]

Thank you for your answer