Parametric LSTM (pLSTM) - a new variant!

Hamid R. Karbasian and Wim M. van Rees

vanRees Lab

Department of Mechanical Engineering

Massachusetts Institute of Technology

lstmNet

LSTM network to predict unsteady flow structures.

step-1:

In the "example" folder, run one of the ode examples. It generates the dataset and corresponding para.txt (including design parameters) in the "data" folder.

step-2:

In the "code" folder, model.py executes training and prediction.

step-3:

In the "results" folder, the trained model and its corresponding files are stored.