[error] [net_module.cpp:37] Net backend not found: onnxruntime #344
Description
I build a docker image with dockerfile-gpu, and follow the instruction in get started. All steps are fine, until the last step, when running the executable object_detection, I get an error like this:
[2022-04-14 01:34:43.313] [mmdeploy] [info] [cuda_device.cpp:84] Default CUDA allocator initialized
[2022-04-14 01:34:43.313] [mmdeploy] [error] [net_module.cpp:37] Net backend not found: onnxruntime
[2022-04-14 01:34:43.313] [mmdeploy] [error] [detector.cpp:58] exception caught: entry not found (8) @ /root/workspace/mmdeploy/csrc/net/net_module.cpp:38
failed to create detector, code: 6
I have already set environment variable PATH and LD_LIBRARY_PATH about onnxruntime lib:
root@5ee45f96fd7b:~/workspace/mmdeploy/build/install/example/build# echo $PATH
/root/workspace/onnxruntime-linux-x64-1.8.1/lib/:/opt/conda/bin:/opt/tensorrt/bin:/usr/local/mpi/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/ucx/bin
root@5ee45f96fd7b:~/workspace/mmdeploy/build/install/example/build# echo $LD_LIBRARY_PATH
/root/workspace/onnxruntime-linux-x64-1.8.1/lib/:/root/workspace/mmdeploy/build/lib:/usr/local/cuda/compat/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
root@5ee45f96fd7b:~/workspace/mmdeploy/build/install/example/build# ls /root/workspace/onnxruntime-linux-x64-1.8.1/lib/
libonnxruntime.so libonnxruntime.so.1.8.1So how can I fix this problem? thanks in advance.
Metadata
Metadata
Assignees
Labels
No labels