search

juripapay / hydronet_parallel

Parallel version of Hydronet
1 stars 0 forks source link

dev02 vs pearl #1

Open juripapay opened 1 year ago

juripapay commented 1 year ago

The parallel version runs on dev02 but not on pearl. There are some issues with importing libraries. The conda environment in both cases is identical:

On dev02 the following import produces no error: from torch_geometric.data import DataListLoader, DataLoader

On pearl there is an error:

from torch_geometric.data import DataLoader Traceback (most recent call last): File "", line 1, in File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/site-packages/torch_geometric/init.py", line 4, in import torch_geometric.data File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/site-packages/torch_geometric/data/init.py", line 1, in from .data import Data File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/site-packages/torch_geometric/data/data.py", line 20, in from torch_sparse import SparseTensor File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/site-packages/torch_sparse/init.py", line 19, in torch.ops.load_library(spec.origin) File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/site-packages/torch/_ops.py", line 643, in load_library ctypes.CDLL(path) File "/mnt/beegfs/home/pearl061/.conda/envs/hydronet_test/lib/python3.8/ctypes/init.py", line 373, in init self._handle = _dlopen(self._name, mode) OSError: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory

juripapay commented 1 year ago

There was a conflict between conda environments. Now the code is running ok.