Hello, I have managed to compile with cuda 10.0, but when I try to train on customed data, it raise error:
Traceback (most recent call last): File "train_net.py", line 57, in <module> main() File "train_net.py", line 49, in main network = make_network(cfg) File "/home/lin/Desktop/clean-pvnet/lib/networks/make_network.py", line 34, in make_network return imp.load_source(module, path).get_network(cfg) File "/home/lin/anaconda3/envs/pvnet/lib/python3.7/imp.py", line 171, in load_source module = _load(spec) File "<frozen importlib._bootstrap>", line 696, in _load File "<frozen importlib._bootstrap>", line 677, in _load_unlocked File "<frozen importlib._bootstrap_external>", line 728, in exec_module File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed File "lib/networks/pvnet/__init__.py", line 1, in <module> from .resnet18 import get_res_pvnet File "lib/networks/pvnet/resnet18.py", line 5, in <module> from lib.csrc.ransac_voting.ransac_voting_gpu import ransac_voting_layer, ransac_voting_layer_v3, estimate_voting_distribution_with_mean File "/home/lin/Desktop/clean-pvnet/lib/csrc/ransac_voting/ransac_voting_gpu.py", line 2, in <module> import lib.csrc.ransac_voting.ransac_voting as ransac_voting ImportError: libcudart.so.9.0: cannot open shared object file: No such file or directory
I'm pretty sure that I have a correct CUDA version setting and environment parameter. I have tried install different version of torch(1.1.0, 1.2.0) and torchvision(0.2.1, 0.4.0), but not work. What could possibly cause it?
Thanks for any help!
Hello, I have managed to compile with cuda 10.0, but when I try to train on customed data, it raise error:
Traceback (most recent call last): File "train_net.py", line 57, in <module> main() File "train_net.py", line 49, in main network = make_network(cfg) File "/home/lin/Desktop/clean-pvnet/lib/networks/make_network.py", line 34, in make_network return imp.load_source(module, path).get_network(cfg) File "/home/lin/anaconda3/envs/pvnet/lib/python3.7/imp.py", line 171, in load_source module = _load(spec) File "<frozen importlib._bootstrap>", line 696, in _load File "<frozen importlib._bootstrap>", line 677, in _load_unlocked File "<frozen importlib._bootstrap_external>", line 728, in exec_module File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed File "lib/networks/pvnet/__init__.py", line 1, in <module> from .resnet18 import get_res_pvnet File "lib/networks/pvnet/resnet18.py", line 5, in <module> from lib.csrc.ransac_voting.ransac_voting_gpu import ransac_voting_layer, ransac_voting_layer_v3, estimate_voting_distribution_with_mean File "/home/lin/Desktop/clean-pvnet/lib/csrc/ransac_voting/ransac_voting_gpu.py", line 2, in <module> import lib.csrc.ransac_voting.ransac_voting as ransac_voting ImportError: libcudart.so.9.0: cannot open shared object file: No such file or directory
I'm pretty sure that I have a correct CUDA version setting and environment parameter. I have tried install different version of torch(1.1.0, 1.2.0) and torchvision(0.2.1, 0.4.0), but not work. What could possibly cause it? Thanks for any help!