Encountering "NYI" problem when convert mask2former to onnx: #3860
Description
I'm still encountering problem after the 4c17c2567075eb17c79123a185d87126a7e87052.patch when convert mask2former to onnx:
Here is my command and env info:
python mmdeploy/tools/deploy.py
mmdeploy/configs/mmseg/segmentation_onnxruntime_dynamic.py
mmsegmentation/configs/mask2former/mask2former_r50_8xb2-160k_ade20k-512x512.py
mmsegmentation/ckpt/mask2former_r50_8xb2-160k_ade20k-512x512_20221204_000055-2d1f55f1.pth
mmsegmentation/demo/demo.png
--work-dir ./
--device cuda
--dump-info
Here is my env settings:
python mmdeploy/tools/check_env.py
08/11 10:41:41 - mmengine - INFO - Environmental information
08/11 10:41:44 - mmengine - INFO - sys.platform: linux
08/11 10:41:44 - mmengine - INFO - Python: 3.9.11 (main, Mar 29 2022, 19:08:29) [GCC 7.5.0]
08/11 10:41:44 - mmengine - INFO - CUDA available: True
08/11 10:41:44 - mmengine - INFO - MUSA available: False
08/11 10:41:44 - mmengine - INFO - numpy_random_seed: 2147483648
08/11 10:41:44 - mmengine - INFO - GPU 0,1: Tesla V100-SXM2-16GB
08/11 10:41:44 - mmengine - INFO - CUDA_HOME: /usr/local/cuda
08/11 10:41:44 - mmengine - INFO - NVCC: Cuda compilation tools, release 12.1, V12.1.105
08/11 10:41:44 - mmengine - INFO - GCC: gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
08/11 10:41:44 - mmengine - INFO - PyTorch: 2.1.0+cu121
08/11 10:41:44 - mmengine - INFO - PyTorch compiling details: PyTorch built with:
- GCC 9.3
- C++ Version: 201703
- Intel(R) oneAPI Math Kernel Library Version 2022.2-Product Build 20220804 for Intel(R) 64 architecture applications
- Intel(R) MKL-DNN v3.1.1 (Git Hash 64f6bcbcbab628e96f33a62c3e975f8535a7bde4)
- OpenMP 201511 (a.k.a. OpenMP 4.5)
- LAPACK is enabled (usually provided by MKL)
- NNPACK is enabled
- CPU capability usage: AVX512
- CUDA Runtime 12.1
- NVCC architecture flags: -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;-gencode;arch=compute_90,code=sm_90
- CuDNN 8.9
- Built with CuDNN 8.9.2
- Magma 2.6.1
- Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=12.1, CUDNN_VERSION=8.9.2, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-invalid-partial-specialization -Wno-unused-private-field -Wno-aligned-allocation-unavailable -Wno-missing-braces -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_DISABLE_GPU_ASSERTS=ON, TORCH_VERSION=2.1.0, 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=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF,
08/11 10:41:44 - mmengine - INFO - TorchVision: 0.16.0+cu121
08/11 10:41:44 - mmengine - INFO - OpenCV: 4.7.0
08/11 10:41:44 - mmengine - INFO - MMEngine: 0.10.7
08/11 10:41:44 - mmengine - INFO - MMCV: 2.1.0
08/11 10:41:44 - mmengine - INFO - MMCV Compiler: GCC 11.4
08/11 10:41:44 - mmengine - INFO - MMCV CUDA Compiler: 12.1
08/11 10:41:44 - mmengine - INFO - MMDeploy: 1.3.1+
08/11 10:41:44 - mmengine - INFO -
08/11 10:41:44 - mmengine - INFO - Backend information
08/11 10:41:44 - mmengine - INFO - tensorrt: None
08/11 10:41:44 - mmengine - INFO - ONNXRuntime: None
08/11 10:41:44 - mmengine - INFO - ONNXRuntime-gpu: 1.19.2
08/11 10:41:44 - mmengine - INFO - ONNXRuntime custom ops: Available
08/11 10:41:44 - mmengine - INFO - pplnn: None
08/11 10:41:44 - mmengine - INFO - ncnn: None
08/11 10:41:44 - mmengine - INFO - snpe: None
08/11 10:41:44 - mmengine - INFO - openvino: None
08/11 10:41:44 - mmengine - INFO - torchscript: 2.1.0+cu121
08/11 10:41:44 - mmengine - INFO - torchscript custom ops: NotAvailable
08/11 10:41:44 - mmengine - INFO - rknn-toolkit: None
08/11 10:41:44 - mmengine - INFO - rknn-toolkit2: None
08/11 10:41:44 - mmengine - INFO - ascend: None
08/11 10:41:44 - mmengine - INFO - coreml: None
08/11 10:41:44 - mmengine - INFO - tvm: None
08/11 10:41:44 - mmengine - INFO - vacc: None
08/11 10:41:44 - mmengine - INFO -
08/11 10:41:44 - mmengine - INFO - Codebase information
08/11 10:41:44 - mmengine - INFO - mmdet: 3.3.0
08/11 10:41:44 - mmengine - INFO - mmseg: 1.2.2
08/11 10:41:44 - mmengine - INFO - mmpretrain: None
08/11 10:41:44 - mmengine - INFO - mmocr: None
08/11 10:41:44 - mmengine - INFO - mmagic: None
08/11 10:41:44 - mmengine - INFO - mmdet3d: None
08/11 10:41:44 - mmengine - INFO - mmpose: None
08/11 10:41:44 - mmengine - INFO - mmrotate: None
08/11 10:41:44 - mmengine - INFO - mmaction: None
08/11 10:41:44 - mmengine - INFO - mmrazor: None
08/11 10:41:44 - mmengine - INFO - mmyolo: None