mmdeploy icon indicating copy to clipboard operation
mmdeploy copied to clipboard

Published 20 hours ago • verified publisher icon open-mmlab

An exception occurred in the inference result of getting mask RCNN with object_detection.py

Open shoayi opened this issue 3 years ago • 3 comments

I use tools/deploy.py to get the TensorRT model and the configuration information is as follows: mmdeploy\configs\mmdet\instance-seg\instance-seg_tensorrt-fp16_static-800x1344.py mmdetection\configs\mask_rcnn\mask_rcnn_x101_64x4d_fpn_mstrain-poly_3x_coco.py mmdetection\checkpoints\mask_rcnn_x101_64x4d_fpn_mstrain-poly_3x_coco_20210526_120447-c376f129.pth mmdetection\demo\demo.jpg --work-dir work_dirs/mask_rcnn/ --device cuda --show --dump-info

i through object_detection.py, resulting in output_detection.png that is different from the example. The mask is displayed in the same shape as the real object, but with a much smaller area.The configuration information about object_detect. py is as follows: cuda mmdetection\mmdeploy\work_dirs\mask_rcnn
mmdetection\demo\demo.jpg

output_detection output_detection

shoayi avatar Aug 31 '22 15:08 shoayi

@lvhan028

tpoisonooo avatar Aug 31 '22 15:08 tpoisonooo

Could you please post the result of python tools/check_env.py?

lvhan028 avatar Sep 01 '22 01:09 lvhan028

Thank you for your reply. The results are as follows: 2022-09-01 09:53:36,393 - mmdeploy - INFO -

2022-09-01 09:53:36,393 - mmdeploy - INFO - Environmental information 2022-09-01 09:53:38,112 - mmdeploy - INFO - sys.platform: win32 2022-09-01 09:53:38,113 - mmdeploy - INFO - Python: 3.7.13 (default, Mar 28 2022, 08:03:21) [MSC v.1916 64 bit (AMD64)] 2022-09-01 09:53:38,113 - mmdeploy - INFO - CUDA available: True 2022-09-01 09:53:38,113 - mmdeploy - INFO - GPU 0: NVIDIA GeForce RTX 3060 2022-09-01 09:53:38,113 - mmdeploy - INFO - CUDA_HOME: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3 2022-09-01 09:53:38,113 - mmdeploy - INFO - NVCC: Cuda compilation tools, release 11.3, V11.3.58 2022-09-01 09:53:38,113 - mmdeploy - INFO - MSVC: 用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.29.30040 版 2022-09-01 09:53:38,113 - mmdeploy - INFO - GCC: n/a 2022-09-01 09:53:38,113 - mmdeploy - INFO - PyTorch: 1.11.0 2022-09-01 09:53:38,113 - mmdeploy - INFO - PyTorch compiling details: PyTorch built with:

  • C++ Version: 199711
  • MSVC 192829337
  • Intel(R) Math Kernel Library Version 2020.0.2 Product Build 20200624 for Intel(R) 64 architecture applications
  • Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e)
  • OpenMP 2019
  • LAPACK is enabled (usually provided by MKL)
  • 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_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 8.2
  • Magma 2.5.4
  • Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=C:/cb/pytorch_1000000000000/work/tmp_bin/sccache-cl.exe, CXX_FLAGS=/DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj -DUSE_PTHREADPOOL -openmp:experimental -IC:/cb/pytorch_1000000000000/work/mkl/include -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DUSE_FBGEMM -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, 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=OFF, USE_NNPACK=OFF, USE_OPENMP=ON, USE_ROCM=OFF,

2022-09-01 09:53:38,113 - mmdeploy - INFO - TorchVision: 0.12.0 2022-09-01 09:53:38,113 - mmdeploy - INFO - OpenCV: 4.6.0 2022-09-01 09:53:38,113 - mmdeploy - INFO - MMCV: 1.6.1 2022-09-01 09:53:38,113 - mmdeploy - INFO - MMCV Compiler: MSVC 192930140 2022-09-01 09:53:38,113 - mmdeploy - INFO - MMCV CUDA Compiler: 11.3 2022-09-01 09:53:38,113 - mmdeploy - INFO - MMDeploy: 0.7.0+18321ee 2022-09-01 09:53:38,113 - mmdeploy - INFO -

2022-09-01 09:53:38,113 - mmdeploy - INFO - Backend information 2022-09-01 09:53:38,736 - mmdeploy - INFO - onnxruntime: 1.8.1 ops_is_avaliable : False 2022-09-01 09:53:38,754 - mmdeploy - INFO - tensorrt: 8.4.3.1 ops_is_avaliable : True 2022-09-01 09:53:38,778 - mmdeploy - INFO - ncnn: None ops_is_avaliable : False 2022-09-01 09:53:38,779 - mmdeploy - INFO - pplnn_is_avaliable: False 2022-09-01 09:53:38,780 - mmdeploy - INFO - openvino_is_avaliable: False 2022-09-01 09:53:38,810 - mmdeploy - INFO - snpe_is_available: False 2022-09-01 09:53:38,810 - mmdeploy - INFO -

2022-09-01 09:53:38,810 - mmdeploy - INFO - Codebase information 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmdet: 2.25.1 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmseg: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmcls: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmocr: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmedit: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmdet3d: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmpose: None 2022-09-01 09:53:38,812 - mmdeploy - INFO - mmrotate: None

Process finished with exit code 0

shoayi avatar Sep 01 '22 01:09 shoayi