Hey maybe I've overlooked something but you guys don't have a rnn encoder with rnn decoder right? So Encode historic + categorical data with lstm_1 to obtain a h and c Run lstm_2 with initialized h and c from encoder other inputs future + categorical data. Possibility to mix encoded h into lstm 2 at the input too, perhaps with a horizon time embedding Possibility to use gru/lstm and recent slstm with improved gating

If you use LSTM with recurrent=True or RNN with recurrent=True or GRU with recurrent=True you get encoder-decoder being recurrent. Your other remarks - feel free to submit a custom architecture, or change it how you like for yourself.

Legend, thanks for the quick reply you guys are great!