error: libonnxruntime.so.1.11.1: cannot open shared object file: No such file or directory

JinqingZhengTju commented 1 year ago

Checklist

[ ] I have searched related issues but cannot get the expected help.
[ ] 2. I have read the FAQ documentation but cannot get the expected help.
[ ] 3. The bug has not been fixed in the latest version.

Describe the bug

When I run inference_model with .onnx file, there is a error as follows: File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/apis/inference.py", line 169, in inference_model model = task_processor.init_backend_model(backend_files) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/mmrotate/deploy/rotated_detection.py", line 101, in init_backend_model model_files, self.model_cfg, self.deploy_cfg, device=self.device) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/mmrotate/deploy/rotated_detection_model.py", line 264, in build_rotated_detection_model kwargs) File "/home/zjq/anaconda3/envs/svd/lib/python3.7/site-packages/mmcv/utils/registry.py", line 237, in build return self.build_func(*args, *kwargs, registry=self) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/mmrotate/deploy/rotated_detection_model.py", line 18, in __build_backend_model return registry.module_dict[cls_name](args, kwargs) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/mmrotate/deploy/rotated_detection_model.py", line 57, in init backend=backend, backend_files=backend_files, device=device) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/mmrotate/deploy/rotated_detection_model.py", line 76, in _init_wrapper deploy_cfg=self.deploy_cfg) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/codebase/base/backend_model.py", line 63, in _build_wrapper output_names=output_names) File "/media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/backend/onnxruntime/wrapper.py", line 46, in init session_options.register_custom_ops_library(ort_custom_op_path) onnxruntime.capi.onnxruntime_pybind11_state.Fail: [ONNXRuntimeError] : 1 : FAIL : Failed to load library /media/zjq/Data/Ubuntu_Project/svd/MMDeploy/mmdeploy/lib/libmmdeploy_onnxruntime_ops.so with error: libonnxruntime.so.1.11.1: cannot open shared object file: No such file or directory

Reproduction

I run this code: model_output = inference_model(model_cfg, deploy_cfg, backend_files, img, device) with the configs: oriented_rcnn_r50_fpn_3x_dota_le90.py rotated-detection_onnxruntime_dynamic.py oriented_rcnn_r50_fpn_1x_dota_le90-6d2b2ce0.onnx

Environment

2022-11-16 22:36:38,887 - mmdeploy - INFO - 

2022-11-16 22:36:38,888 - mmdeploy - INFO - **********Environmental information**********
2022-11-16 22:36:39,045 - mmdeploy - INFO - sys.platform: linux
2022-11-16 22:36:39,045 - mmdeploy - INFO - Python: 3.7.13 (default, Oct 18 2022, 18:57:03) [GCC 11.2.0]
2022-11-16 22:36:39,045 - mmdeploy - INFO - CUDA available: True
2022-11-16 22:36:39,045 - mmdeploy - INFO - GPU 0: NVIDIA GeForce RTX 3090
2022-11-16 22:36:39,045 - mmdeploy - INFO - CUDA_HOME: /usr/local/cuda
2022-11-16 22:36:39,045 - mmdeploy - INFO - NVCC: Cuda compilation tools, release 11.3, V11.3.109
2022-11-16 22:36:39,045 - mmdeploy - INFO - GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
2022-11-16 22:36:39,045 - mmdeploy - INFO - PyTorch: 1.12.1
2022-11-16 22:36:39,045 - mmdeploy - INFO - PyTorch compiling details: PyTorch built with:
  - GCC 9.3
  - C++ Version: 201402
  - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: AVX2
  - CUDA Runtime 11.3
  - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37
  - CuDNN 8.3.2  (built against CUDA 11.5)
  - Magma 2.5.2
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.3.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -fabi-version=11 -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.12.1, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, 

2022-11-16 22:36:39,045 - mmdeploy - INFO - TorchVision: 0.13.1
2022-11-16 22:36:39,045 - mmdeploy - INFO - OpenCV: 4.1.2
2022-11-16 22:36:39,045 - mmdeploy - INFO - MMCV: 1.6.0
2022-11-16 22:36:39,045 - mmdeploy - INFO - MMCV Compiler: GCC 9.3
2022-11-16 22:36:39,045 - mmdeploy - INFO - MMCV CUDA Compiler: 11.3
2022-11-16 22:36:39,045 - mmdeploy - INFO - MMDeploy: 0.9.0+4c872a4
2022-11-16 22:36:39,045 - mmdeploy - INFO - 

2022-11-16 22:36:39,045 - mmdeploy - INFO - **********Backend information**********
2022-11-16 22:36:40,511 - mmdeploy - INFO - onnxruntime: 1.11.1 ops_is_avaliable : True
2022-11-16 22:36:40,558 - mmdeploy - INFO - tensorrt: 8.2.3.0   ops_is_avaliable : True
2022-11-16 22:36:40,580 - mmdeploy - INFO - ncnn: None  ops_is_avaliable : False
2022-11-16 22:36:40,582 - mmdeploy - INFO - pplnn_is_avaliable: False
2022-11-16 22:36:40,584 - mmdeploy - INFO - openvino_is_avaliable: False
2022-11-16 22:36:40,607 - mmdeploy - INFO - snpe_is_available: False
2022-11-16 22:36:40,609 - mmdeploy - INFO - ascend_is_available: False
2022-11-16 22:36:40,611 - mmdeploy - INFO - coreml_is_available: False
2022-11-16 22:36:40,611 - mmdeploy - INFO - 

2022-11-16 22:36:40,611 - mmdeploy - INFO - **********Codebase information**********
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmdet:  2.25.2
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmseg:  0.29.0
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmcls:  0.24.0
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmocr:  None
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmedit: None
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmdet3d:    None
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmpose: None
2022-11-16 22:36:40,614 - mmdeploy - INFO - mmrotate:   0.3.2

Error traceback

No response

adrigrillo commented 1 year ago

Hello,

Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.

export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH

That solved the problem for me last time.

JinqingZhengTju commented 1 year ago

Hello,

Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.

I have set the environment path for onnxruntime as same as your advice. But it also reports this error, I don't know the reason.

xin-li-67 commented 1 year ago

Hello, Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.
I have set the environment path for onnxruntime as same as your advice. But it also reports this error, I don't know the reason.

Is this issue solved? If not, would you please attach a screenshot of your echo $LD_LIBRARY_PATH ?

JinqingZhengTju commented 1 year ago

Hello, Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.
I have set the environment path for onnxruntime as same as your advice. But it also reports this error, I don't know the reason.
Is this issue solved? If not, would you please attach a screenshot of your echo $LD_LIBRARY_PATH ?

I have solved this issue by setting the environment path in Pycharm. This is a way to solve this kind of problem. I don't know why the environment path setting in ~./bashrc doesn't work.

xin-li-67 commented 1 year ago

Hello, Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.
I have set the environment path for onnxruntime as same as your advice. But it also reports this error, I don't know the reason.
Is this issue solved? If not, would you please attach a screenshot of your echo $LD_LIBRARY_PATH ?
I have solved this issue by setting the environment path in Pycharm. This is a way to solve this kind of problem. I don't know why the environment path setting in ~./bashrc doesn't work.

Glad to hear that! BTW, have you done source ~/.bashrc after you set the path in ~/.bashrc ???

JinqingZhengTju commented 1 year ago

Hello, Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.
I have set the environment path for onnxruntime as same as your advice. But it also reports this error, I don't know the reason.
Is this issue solved? If not, would you please attach a screenshot of your echo $LD_LIBRARY_PATH ?
I have solved this issue by setting the environment path in Pycharm. This is a way to solve this kind of problem. I don't know why the environment path setting in ~./bashrc doesn't work.
Glad to hear that! BTW, have you done source ~/.bashrc after you set the path in ~/.bashrc ???

Yes, I have run this command. But it still doesn't work. I have to set it in Pycharm.

lvhan028 commented 1 year ago

If you echo $LD_LIBRARY_PATH in another shell, what will print? Have you tried to reopen PyCharm?

NEAR07 commented 7 months ago

Hello,

Try to set ONNXRUNTIME_DIR with the location where you have installed the onnx-runtime changing the following lines.
export ONNXRUNTIME_DIR="path to onnx-runtime"
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
That solved the problem for me last time.

where do you put the library onnx-runetime?

open-mmlab / mmdeploy