I am interested in applying recurrent neutral with a bi-LSTM network for time series data. I believe that
Moving Time Window: Taylor Green Vortex Decay example can be useful for this purpose. it is possible to use Moving Time Window front and back in time to achieve the behavior of bi-LSTM. or there is any other way to achieve it, thank you
We currently do not natively support RNNs inside Modulus-Sym, but we encourage our users to customize and experiment since this is an active topic of research after all. Good luck!