seamseg icon indicating copy to clipboard operation
seamseg copied to clipboard

ImportError: libtorch.so: cannot open shared object file: No such file or directory

Open otoofim opened this issue 4 years ago • 1 comments

Hi,

I am trying to load the pertained mode provided by you to carry out some evaluation on a set of images. I followed the instruction you provided to install the dependencies and packages. However, I am getting the error below which implies that a package is missing. I appreciate if you could advise me on this issue.

""""""""""""""""""""""""""""""""""""""""""""""" Traceback (most recent call last): File "test_panoptic.py", line 17, in from seamseg.algos.detection import PredictionGenerator as BbxPredictionGenerator, DetectionLoss,
File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/site-packages/seamseg-0.1.dev37+g3d10aea-py3.6-linux-x86_64.egg/seamseg/algos/detection.py", line 6, in from seamseg.utils.bbx import ious, calculate_shift, bbx_overlap, mask_overlap File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/site-packages/seamseg-0.1.dev37+g3d10aea-py3.6-linux-x86_64.egg/seamseg/utils/bbx/init.py", line 1, in from .bbx import * File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/site-packages/seamseg-0.1.dev37+g3d10aea-py3.6-linux-x86_64.egg/seamseg/utils/bbx/bbx.py", line 5, in from . import _backend ImportError: libtorch.so: cannot open shared object file: No such file or directory Traceback (most recent call last): File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/site-packages/torch/distributed/launch.py", line 235, in main() File "PATH-TO-ANACONDA/envs/seamseg/lib/python3.6/site-packages/torch/distributed/launch.py", line 231, in main cmd=process.args) subprocess.CalledProcessError: Command '['PATH-TO-ANACONDA/envs/seamseg/bin/python', '-u', 'test_panoptic.py', '--local_rank=0', '--meta', '../checkpoints/seamseg_r50_vistas/metadata.bin', '--log_dir', '../log', '../checkpoints/seamseg_r50_vistas/config.ini', '../checkpoints/seamseg_r50_vistas/seamseg_r50_vistas.tar', '../../signals/data/frames/23', '../out']' returned non-zero exit status 1.

"""""""""""""""""""""""""""""""""""""""""""""""

My system config is : CUDA10.1 GCC7 Python 3.6 torch1.1

I ran this command to do inference:

python -m torch.distributed.launch --nproc_per_node=1 test_panoptic.py --meta "../checkpoints/seamseg_r50_vistas/metadata.bin" --log_dir '../log' '../checkpoints/seamseg_r50_vistas/config.ini' '../checkpoints/seamseg_r50_vistas/seamseg_r50_vistas.tar' '../frames' '../out'

otoofim avatar Jun 14 '21 13:06 otoofim

I've ran into the same error, have you solved the problem?

youssef007 avatar Oct 12 '21 16:10 youssef007