PINO

PINO copied to clipboard

I realise that for the Physics informed part you use MSE Loss but for the pointwise solution error (supervised training) you use your custom LP Loss function. I understand that...

cfos3120

PDE Loss makes the result worse

1

Thanks for your great job! Recently I am trying to reimplement the PINO on Darcy Flow. I found that if I set `f_loss=0`, the result is getting better and converge...

JiahaoHuang99

This pull request fixes bugs in the Darcy Flow pre-training example with PINO with `train_operator.py` and `Darcy-pretrain.yaml`. some keys were missing in the config file and some variables were named...

snawarhussain

Hi, I need to generate longer Burger trajectories. I think found the MATLAB files to solve the Burgers equation (I assume you follow this code https://github.com/PredictiveIntelligenceLab/Physics-informed-DeepONets/tree/main/Burger/Data). Can you please share...

federicobergamin

I am Junho, a researcher who is studying about PINN at SungKun Korea. At my paper, I am going to write and refer your results about Kolmogorov flows implemented by...

cjunho

Vorticity to velocity

1

Hello, thank you for you work. I have a misunderstanding with the obtaining velocity (for Navier Stokes equation) from given vorticity part. We actually need it for loss and you...

GerasimovIV

Could you please provide the demo code for inverse problems? Thanks

Shancong-Mou

link to Burgers mat file : Access denied

1

elhamod

In line 131, code `index_t = torch.zeros(nx,).long()` should be modified to `index_t = torch.zeros(nt,).long()`

xubonan

FD-based gradient calculation seems incorrect for Burgers (with code to verify)

2

# The issue I am trying to understand how gradients are computed for Burgers, implemented by `FDM_Burgers()` in `train_utils/losses.py`, as pasted below: ```python def FDM_Burgers(u, v, D=1): batchsize = u.size(0)...

kuangdai