Error using trainNetwork when having 3 parameter in X and one in Y

The training sequences are of feature dimension 472 but the input layer expects sequences of

feature dimension 3.

Pls tell me how i can stop getting this error.

[David Ho]

Hello Deepak,

Are you able to share an example of code that reproduces this error? This would help the community here to try and work out how to resolve the issue.

Answer 1

[Swaraj]

Hi,

I understand your problem and go through the possible reasons for it.

The dimensions of the input sequences that your neural network model anticipates do not coincide with the dimensions of your training data, which results in this error. Your input sequence's time steps each have 472 features because the training data's feature dimension is 472. Nevertheless, because your input layer is set up to anticipate sequences of 3 characteristics, each time step in the input sequence must have 3 features.

Go through the following documentation for understanding trainNetwork.

Train deep learning neural network - MATLAB trainNetwork (mathworks.com)

I hope it helps.

Thanks!!