Code for Darcy Flow does not work

[Shancong-Mou]

Hi,

Amazing project! I am able to duplicate the Burgers experiment. However, the code for Darcy Flow does not work.

python3 train_operator.py --config_path configs/pretrain/Darcy-pretrain.yaml

with the following error:
Traceback (most recent call last):
File "train_operator.py", line 133, in
train_2d(args, config)
File "train_operator.py", line 90, in train_2d
num=data_config['n_samples'],
KeyError: 'n_samples'

[alabaykazakh]

Hi,

Amazing project! I am able to duplicate the Burgers experiment. However, the code for Darcy Flow does not work.

python3 train_operator.py --config_path configs/pretrain/Darcy-pretrain.yaml

with the following error: Traceback (most recent call last): File "train_operator.py", line 133, in train_2d(args, config) File "train_operator.py", line 90, in train_2d num=data_config['n_samples'], KeyError: 'n_samples'

write n_samples instead of n_sample in configs/pretrain/Darcy-pretrain.yaml file.

[Shancong-Mou]

Hi,
Amazing project! I am able to duplicate the Burgers experiment. However, the code for Darcy Flow does not work.
python3 train_operator.py --config_path configs/pretrain/Darcy-pretrain.yaml
with the following error: Traceback (most recent call last): File "train_operator.py", line 133, in train_2d(args, config) File "train_operator.py", line 90, in train_2d num=data_config['n_samples'], KeyError: 'n_samples'

write n_samples instead of n_sample in configs/pretrain/Darcy-pretrain.yaml file.

Thanks for your feedback. I just fix that issue and compiled again, a new error appears:

Traceback (most recent call last):
File "train_operator.py", line 133, in
train_2d(args, config)
File "train_operator.py", line 98, in train_2d
pad_ratio=config['model']['pad_ratio']).to(device)
KeyError: 'pad_ratio'

I am wondering if the author accidentally push the run version of the code?

[devzhk]

Hey,

Sorry, my bad. I added a new feature to the FNO architecture (padding) and didn't merge it correctly with the main branch.

I just updated the repo. For Darcy exp, you can run the following to train the operator

python3 train_darcy.py --config configs/operator/Darcy-pretrain.yaml

To test the trained operator on test set, run

python3 train_darcy.py --config configs/operator/Darcy-pretrain.yaml --test --ckpt [path to the checkpoint]

Regards