yuzhiguo07 / protein-structure-embedding Goto Github PK

Hello,
I encountered a ModuleNotFoundError while trying to run the build_dataset_sample.py script in the project.
Here's the error message I received:

python build_dataset_sample.py
Using backend: pytorch
Traceback (most recent call last):
File "build_dataset_sample.py", line 14, in
from tfold_se3.utils import tfold_init
File "/userhome/aimia/lyt/Protein-Structure-Embedding/tfold_se3/utils/init.py", line 28, in
from tfold_se3.utils.se3_utils import get_basis_and_radial
File "/userhome/aimia/lyt/Protein-Structure-Embedding/tfold_se3/utils/se3_utils.py", line 9, in
from tfold_se3.from_se3cnn import utils_steerable
ModuleNotFoundError: No module named 'tfold_se3.from_se3cnn'

It seems that there might be an issue with the project structure or a missing file. Could you please help me resolve this issue? Any guidance on how to fix this problem would be greatly appreciated.

Thank you.

Since the data preprocessing part of the work was completed in cooperation with co-workers during the internship, the pdb_parser has not yet been authorized for public release. Therefore, I just attached the backbone model for sampling embedding without the data preprocessing tools. You may create your pdb_parser using Bio package in python and get the hdf5.files.