search

lululxvi / deeponet

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

The data shape of the triplet if the number of input functions of branch nets is more than one? #9

Closed WangYicunZJU closed 2 years ago

WangYicunZJU commented 2 years ago

Hello: First of all, thank you very much for your excellent work! I have met a little problem. How to reshape the dataset (or change the code) if the number of input functions of branch nets is more than one? Just like G_phi in this paper("DeepM&Mnet: Inferring the electroconvection multiphysics fields..."), (c+, c-) -->phi. Or like G_rhoNO in this paper("DeepM&Mnet for hypersonics..."), (rho_N2, rho_O2) --> rho_NO. Specifically, what about the data shape of the triplet? Thanks in advance.

lululxvi commented 2 years ago

Good question. In the DeepMMnet paper, we simply concatenate the two functions into a long vector.

WangYicunZJU commented 2 years ago

Hello Dr. Lu, It is very nice to get your reply. Following your answer, I think the data of the branch nets may be like this: [ u1(x1), ..., u1(xm) ] for one input function u1; [ u1(x1), ..., u1(xm), u2(x1), ..., u2(xm) ] for two input function u1 & u2. Or like this: [ u1(x1), u2(x1),..., u1(xm), u2(xm) ] ? I think the former seems to be the case. Do I have the correct understanding? I am looking forward to your reply. Thanks in advance.

lululxvi commented 2 years ago

Yes, we use the former.

WangYicunZJU commented 2 years ago

OK, Dr. Lu. I think I have got it. Thanks again!