Closed HydrogenSulfate closed 7 months ago
Thank you for pointing this out! It is a typo and we have fixed it.
Thank you for pointing this out! It is a typo and we have fixed it.
And I've found another question, is the run shell script you given correct? Cuz when I run python3 main.py --config/sota.py
, error occurred below:
2024-04-09 15:47:16.206208: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
/jax/lib/python3.9/site-packages/absl/flags/_validators.py:254: UserWarning: Flag --config has a non-None default value; therefore, mark_flag_as_required will pass even if flag is not specified in the command line!
mark_flag_as_required(flag_name, flag_values)
/jax/lib/python3.9/site-packages/absl/flags/_validators.py:254: UserWarning: Flag --workdir has a non-None default value; therefore, mark_flag_as_required will pass even if flag is not specified in the command line!
mark_flag_as_required(flag_name, flag_values)
FATAL Flags parsing error: Unknown command line flag 'config/sota.py'
Pass --helpshort or --helpfull to see help on flags.
I guess the correct running script is: python3 main.py --config configs/sota.py
, which can run properly.
If that, all the scripts in your README.md should be correct.
Thank you for raising this. We apologize for any confusion caused. The correct command to specify different configs is
python3 main.py --config=configs/sota.py
By default, it will use default
config, so users can directly runpython3 main.py
.
We've included an additional example command to guide users in specifying different configs.
Thank you for raising this. We apologize for any confusion caused. The correct command to specify different configs is
python3 main.py --config=configs/sota.py
By default, it will use
default
config, so users can directly runpython3 main.py
.We've included an additional example command to guide users in specifying different configs.
There seems still 3 scripts to be corrected within 3 example directory: https://github.com/search?q=repo%3APredictiveIntelligenceLab%2Fjaxpi+%22--config%2F%22&type=code
Jaxpi is a great work for assembling several effective and practical PINN related methods. But according to wandb you given, L2 error of Allen_cahn should be $5.174\times10^{-1}$ rather than $51.74\times 10^{-1}$(that's a really exaggerative error 😲 )?
https://github.com/PredictiveIntelligenceLab/jaxpi/blob/c91a26c0c410553a81690552e1c3d1f04cafea51/examples/allen_cahn/README.md?plain=1#L29