lululxvi / deeponet

Learning nonlinear operators via DeepONet
Other
474 stars 130 forks source link

How advanced DeepOnet can be compatible with simple Fully-connected networks #41

Open LVJIAQI777 opened 10 months ago

LVJIAQI777 commented 10 months ago

Hello Dr. Lu, Thank you very much for the DeepXDE you developed. I would like to ask you a question about DeepOnets. It is reasonable that advanced DeepOnet can be compatible with simple FNN. For example, in a simple house price prediction problem, we have 2 inputs, like the size and age of the house. Then, we want to predict the price of the house. There seems to be no trunk net counterpart in this example. So, how do we implement a simple FNN with DeepOnet? I do know that I can use FNN directly. I'm still trying to figure out how to be compatible with a simple network using DeepOnet architecture. Could you please answer my confusion? I am looking forward to your reply!

lululxvi commented 10 months ago

In this case, FNN seems good enough. Why do you have to use DeepONet?