We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
mmdeploy1.3.1,导出faster-rcnn hr18的onnx模型后,用onnxruntime跑推理,出现没法和pytorch对齐问题。像素偏差5个像素。
https://github.com/open-mmlab/mmdetection/tree/main/configs/hrnet
(mmdeploy) F:\youtu\ocr\test\PaddleOCR-plus>python F:\youtu\sdk\mmdeploy\tools/check_env.py 11/14 20:22:07 - mmengine - INFO - 11/14 20:22:07 - mmengine - INFO - **********Environmental information********** 11/14 20:22:19 - mmengine - INFO - sys.platform: win32 11/14 20:22:19 - mmengine - INFO - Python: 3.9.19 (main, May 6 2024, 20:12:36) [MSC v.1916 64 bit (AMD64)] 11/14 20:22:19 - mmengine - INFO - CUDA available: False 11/14 20:22:19 - mmengine - INFO - MUSA available: False 11/14 20:22:19 - mmengine - INFO - numpy_random_seed: 2147483648 11/14 20:22:19 - mmengine - INFO - MSVC: 用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.39.33523 版 11/14 20:22:19 - mmengine - INFO - GCC: n/a 11/14 20:22:19 - mmengine - INFO - PyTorch: 1.12.1 11/14 20:22:19 - mmengine - 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 v2.6.0 (Git Hash 52b5f107dd9cf10910aaa19cb47f3abf9b349815) - OpenMP 2019 - LAPACK is enabled (usually provided by MKL) - CPU capability usage: AVX2 - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, 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=1.12.1, USE_CUDA=0, USE_CUDNN=OFF, 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, 11/14 20:22:19 - mmengine - INFO - TorchVision: 0.13.1 11/14 20:22:19 - mmengine - INFO - OpenCV: 4.10.0 11/14 20:22:19 - mmengine - INFO - MMEngine: 0.10.4 11/14 20:22:19 - mmengine - INFO - MMCV: 2.0.1 11/14 20:22:19 - mmengine - INFO - MMCV Compiler: MSVC 192930148 11/14 20:22:19 - mmengine - INFO - MMCV CUDA Compiler: not available 11/14 20:22:19 - mmengine - INFO - MMDeploy: 1.3.1+255690a 11/14 20:22:19 - mmengine - INFO - 11/14 20:22:19 - mmengine - INFO - **********Backend information********** 11/14 20:22:19 - mmengine - INFO - tensorrt: None 11/14 20:22:20 - mmengine - INFO - ONNXRuntime: 1.19.2 11/14 20:22:20 - mmengine - INFO - ONNXRuntime-gpu: None 11/14 20:22:20 - mmengine - INFO - ONNXRuntime custom ops: Available 11/14 20:22:20 - mmengine - INFO - pplnn: None 11/14 20:22:20 - mmengine - INFO - ncnn: None 11/14 20:22:20 - mmengine - INFO - snpe: None 11/14 20:22:20 - mmengine - INFO - openvino: None 11/14 20:22:20 - mmengine - INFO - torchscript: 1.12.1 11/14 20:22:20 - mmengine - INFO - torchscript custom ops: NotAvailable 11/14 20:22:20 - mmengine - INFO - rknn-toolkit: None 11/14 20:22:20 - mmengine - INFO - rknn-toolkit2: None 11/14 20:22:20 - mmengine - INFO - ascend: None 11/14 20:22:20 - mmengine - INFO - coreml: None 11/14 20:22:20 - mmengine - INFO - tvm: None 11/14 20:22:20 - mmengine - INFO - vacc: None 11/14 20:22:20 - mmengine - INFO - 11/14 20:22:20 - mmengine - INFO - **********Codebase information********** 11/14 20:22:20 - mmengine - INFO - mmdet: 3.1.0 11/14 20:22:20 - mmengine - INFO - mmseg: None 11/14 20:22:20 - mmengine - INFO - mmpretrain: 1.2.0 11/14 20:22:20 - mmengine - INFO - mmocr: None 11/14 20:22:20 - mmengine - INFO - mmagic: None 11/14 20:22:20 - mmengine - INFO - mmdet3d: None 11/14 20:22:20 - mmengine - INFO - mmpose: None 11/14 20:22:20 - mmengine - INFO - mmrotate: None 11/14 20:22:20 - mmengine - INFO - mmaction: None 11/14 20:22:20 - mmengine - INFO - mmrazor: None 11/14 20:22:20 - mmengine - INFO - mmyolo: None
No response
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Checklist
Describe the bug
mmdeploy1.3.1,导出faster-rcnn hr18的onnx模型后,用onnxruntime跑推理,出现没法和pytorch对齐问题。像素偏差5个像素。
Reproduction
https://github.com/open-mmlab/mmdetection/tree/main/configs/hrnet
Environment
Error traceback
No response
The text was updated successfully, but these errors were encountered: