BiLSTM layer prediction and target mismatch error.

I am trying to do a sequence to sequence modelling using BilSTM network. The data is arranged for inputs as cells where each cell corresponds to an input sequence with TxC elements per cell, where T is the number of time steps and C is the number of features. The output sequence is arranaged in cells where each cell corresponds to an output sequence with TxC elements per cell.

The BilSTM layer is followed by a fullyconnectedlayer (FCL), where the output of the FCL is set as 16. I am using the trainnet function. When I am running the code it gives me an error that the size of predictions and targets must match. What can I do to resolve the issue?

Answer 1

[Taylor]

Hard to give advice more specific than the error message without your code for context. You should check out this example for reference.