Tensorrt model inference with mmdetection?

[a227799770055]

Hi guys,
I have a question about model inference.
After I successfully converted PyTorch model to tensorrt model. How can I inference image with the new model?
Does the project of MMDetection have api for us to use it? Or there have any tutorial can help me.

Thanks

[RunningLeon]

@a227799770055 Hi, you can refer to the tutorial: how_to_measure_performance_of_models.

[a227799770055]

@RunningLeon Thanks for your answer.
However, when I run the test script encounter the following error.
How can I fix the problem?

File "tools/test.py", line 138, in <module> main() File "tools/test.py", line 130, in main outputs = task_processor.single_gpu_test(model, data_loader, args.show, File "/home/insign/Doc/insign/mmdeploy/mmdeploy/codebase/base/task.py", line 137, in single_gpu_test return self.codebase_class.single_gpu_test(model, data_loader, show, File "/home/insign/Doc/insign/mmdeploy/mmdeploy/codebase/mmdet/deploy/mmdetection.py", line 142, in single_gpu_test outputs = single_gpu_test(model, data_loader, show, out_dir, **kwargs) File "/home/insign/.local/lib/python3.8/site-packages/mmdet/apis/test.py", line 65, in single_gpu_test result = [(bbox_results, encode_mask_results(mask_results)) File "/home/insign/.local/lib/python3.8/site-packages/mmdet/apis/test.py", line 65, in <listcomp> result = [(bbox_results, encode_mask_results(mask_results)) File "/home/insign/.local/lib/python3.8/site-packages/mmdet/core/mask/utils.py", line 58, in encode_mask_results np.array( File "/home/insign/.local/lib/python3.8/site-packages/torch/_tensor.py", line 680, in __array__ return self.numpy().astype(dtype, copy=False) TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

[RunningLeon]

@a227799770055 Hi, could you post here what the script you ran and output of python tools/check_env.py here?

[a227799770055]

@RunningLeon thanks for your reply.

Script
!python tools/test.py /home/insign/Doc/insign/mmdeploy/configs/mmdet/instance-seg/instance-seg_tensorrt-int8_static-800x1344.py /home/insign/Doc/insign/mmdeploy/nuclei_custom_config.py --model /home/insign/Doc/insign/mmdeploy/work_dir/end2end.engine --out out.pkl --device cuda:0

check_env
2022-04-19 11:30:58,732 - mmdeploy - INFO -

2022-04-19 11:30:58,732 - mmdeploy - INFO - Environmental information
2022-04-19 11:30:59,490 - mmdeploy - INFO - sys.platform: linux
2022-04-19 11:30:59,490 - mmdeploy - INFO - Python: 3.8.10 (default, Mar 15 2022, 12:22:08) [GCC 9.4.0]
2022-04-19 11:30:59,490 - mmdeploy - INFO - CUDA available: True
2022-04-19 11:30:59,490 - mmdeploy - INFO - GPU 0: NVIDIA GeForce RTX 3090
2022-04-19 11:30:59,490 - mmdeploy - INFO - CUDA_HOME: /usr/local/cuda-11.3
2022-04-19 11:30:59,490 - mmdeploy - INFO - NVCC: Build cuda_11.3.r11.3/compiler.29745058_0
2022-04-19 11:30:59,490 - mmdeploy - INFO - GCC: gcc (Ubuntu 7.5.0-6ubuntu2) 7.5.0
2022-04-19 11:30:59,490 - mmdeploy - INFO - PyTorch: 1.10.2+cu113
2022-04-19 11:30:59,490 - mmdeploy - INFO - PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
• Intel(R) MKL-DNN v2.2.3 (Git Hash 7336ca9f055cf1bfa13efb658fe15dc9b41f0740)
• OpenMP 201511 (a.k.a. OpenMP 4.5)
• LAPACK is enabled (usually provided by MKL)
• NNPACK is enabled
• CPU capability usage: AVX512
• 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_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
• CuDNN 8.2
• Magma 2.5.2
• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -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-sign-compare -Wno-unused-parameter -Wno-unused-variable -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 -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.10.2, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

2022-04-19 11:30:59,490 - mmdeploy - INFO - TorchVision: 0.11.3+cu113
2022-04-19 11:30:59,490 - mmdeploy - INFO - OpenCV: 4.5.5
2022-04-19 11:30:59,490 - mmdeploy - INFO - MMCV: 1.4.6
2022-04-19 11:30:59,490 - mmdeploy - INFO - MMCV Compiler: GCC 7.3
2022-04-19 11:30:59,490 - mmdeploy - INFO - MMCV CUDA Compiler: 11.3
2022-04-19 11:30:59,490 - mmdeploy - INFO - MMDeploy: 0.4.0+85c46ee
2022-04-19 11:30:59,490 - mmdeploy - INFO -

2022-04-19 11:30:59,490 - mmdeploy - INFO - Backend information
2022-04-19 11:30:59,646 - mmdeploy - INFO - onnxruntime: 1.10.0 ops_is_avaliable : True
2022-04-19 11:30:59,646 - mmdeploy - INFO - tensorrt: 8.0.1.6 ops_is_avaliable : True
2022-04-19 11:30:59,647 - mmdeploy - INFO - ncnn: None ops_is_avaliable : False
2022-04-19 11:30:59,647 - mmdeploy - INFO - pplnn_is_avaliable: False
2022-04-19 11:30:59,648 - mmdeploy - INFO - openvino_is_avaliable: False
2022-04-19 11:30:59,648 - mmdeploy - INFO -

2022-04-19 11:30:59,648 - mmdeploy - INFO - Codebase information
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmdet: 2.22.0
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmseg: None
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmcls: 0.20.1
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmocr: None
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmedit: None
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmdet3d: None
2022-04-19 11:30:59,649 - mmdeploy - INFO - mmpose: None

[RunningLeon]

@a227799770055 Hi, this has been fixed in #276 , pls. kindly use master branch or released v0.4.0.

[a227799770055]

@RunningLeon Thanks for your help!