noqa: F403 _ Something went wrong with running generating augmentation data. Exiting.

[Hedi65]

Dear developer

i am getting this error and have no idea why, all the paths are absolute as well.

CUDA_VISIBLE_DEVICES=0 bash run_comebin.sh -t 12 -a /home/user/Documents/HIAS/fasta_files/Zone01.fa -p /home/user/Documents/HIAS/BAM/1 -o /home/user/Documents/HIAS/Comebin_res/Z01

Batch size: 1024
N50: 1738
Tau(temperature): 0.15
/home/user/Documents/HIAS/Comebin_res/Z01/data_augmentation does not exist.
Running data augmentation.
Traceback (most recent call last):
File "main.py", line 7, in
from train_CLmodel import train_CLmodel
File "/home/user/miniconda3/envs/comebin/bin/COMEBin/train_CLmodel.py", line 1, in
import torch
File "/home/user/miniconda3/envs/comebin/lib/python3.7/site-packages/torch/init.py", line 197, in
from torch._C import * # noqa: F403
ImportError: /home/user/miniconda3/envs/comebin/lib/python3.7/site-packages/torch/lib/libtorch_cpu.so: undefined symbol: iJIT_NotifyEvent
Something went wrong with running generating augmentation data. Exiting.

thanks for your help

[zhengzhengzhao468-del]

Have you solved this problem? I have also encountered the same issue.

[Hedi65]

no specific reason, really. I just restarted the system, and the problem was solved. Hope this helps you

[zhengzhengzhao468-del]

Thank you for your response. I believe my reason is that the system GPU and PyTorch do not match.

[wdp9966]

This symbol was removed in MKL 2024.1, but PyTorch is built against an older version of MKL.
It is recommended to run "pip install mkl==2024.0"