Let’s assume I have “abbcab”-> {“Apple”,”Banana”}, “Bibb”-> {“Carrot”,“Padding”}. where the feature is a sequence of characters, so a string, and the target is a list of words. The target maximum length is 11, and the maximum feature length is 208. While building an RNN, I just set input-> NetEncoder[{“Characters”}], but I want to output-> NetDecoder[{“Tokens”, targetVal}]. However, I don’t know how to set up the last layer to get such output. I did check the tutorial/NeuralNetworksSequenceLearning##1680168479
You are referring to the right tutorial page (tutorial/NeuralNetworksSequenceLearning#1680168479 : subsection “Integer Addition with Variable-Length Output” of “Sequence-to-Sequence Learning”).
So you should also define an architecture in two parts: an encoderNet, which encodes your input string, and a decoderNet which generates the next output token given the previous generated ones (starting with a special token which means “nothing generated yet”, and that can generate a special token to stop the generation).
That is the way to generate variable-length sequences with neural networks. There is no special output layer that can generate sequences of different lengths.