[Bug] bug in convert mmrotate model to onnx

[ZeroLelouch1205]

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 user mmdeploy/tools/depoly.py convert my mmrotate model to onnx,it report 2 warning,it makes my inference result isn't same as pth
like this:

05/08 15:26:01 - mmengine - WARNING - Failed to search registry with scope "mmrotate" in the "Codebases" registry tree. As a workaround, the current "Codebases" registry in "mmdeploy" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmrotate" is a correct scope, or whether the registry is initialized.
05/08 15:26:01 - mmengine - WARNING - Failed to search registry with scope "mmrotate" in the "mmrotate_tasks" registry tree. As a workaround, the current "mmrotate_tasks" registry in "mmdeploy" is used to build instance. This may cause unexpected failure when running the built modules. Please check whether "mmrotate" is a correct scope, or whether the registry is initialized.

and it also report 4 warning:

[W C:\cb\pytorch_1000000000000\work\torch\csrc\jit\passes\onnx\shape_type_inference.cpp:1924] Warning: The shape inference of mmdeploy::MMCVRoIAlignRotated type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function. (function UpdateReliable)
[W C:\cb\pytorch_1000000000000\work\torch\csrc\jit\passes\onnx\shape_type_inference.cpp:1924] Warning: The shape inference of mmdeploy::MMCVRoIAlignRotated type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function. (function UpdateReliable)
[W C:\cb\pytorch_1000000000000\work\torch\csrc\jit\passes\onnx\shape_type_inference.cpp:1924] Warning: The shape inference of mmdeploy::MMCVRoIAlignRotated type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function. (function UpdateReliable)
[W C:\cb\pytorch_1000000000000\work\torch\csrc\jit\passes\onnx\shape_type_inference.cpp:1924] Warning: The shape inference of mmdeploy::NMSRotated type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function. (function UpdateReliable)

Reproduction

my_depoly_config:

backend_config=dict(type='onnxruntime'),
codebase_config=dict(
    type='mmrotate',
    task='RotatedDetection',
    post_processing=dict(
        score_threshold=0.05,
        iou_threshold=0.1,
        pre_top_k=3000,
        keep_top_k=2000,
        max_output_boxes_per_class=2000)
),
onnx_config = dict(
    type='onnx',
    export_params=True,
    keep_initializers_as_inputs=False,
    output_names=['dets', 'labels'],
    input_names=['input'],
    input_shape=[1024, 1024],
    opset_version=11,
    optimize=True
)

Environment

05/08 15:24:09 - mmengine - INFO - 

05/08 15:24:09 - mmengine - INFO - **********Environmental information**********
05/08 15:24:12 - mmengine - INFO - sys.platform: win32
05/08 15:24:12 - mmengine - INFO - Python: 3.9.22 | packaged by conda-forge | (main, Apr 14 2025, 23:26:18) [MSC v.1929 64 bit (AMD64)]
05/08 15:24:12 - mmengine - INFO - CUDA available: True
05/08 15:24:12 - mmengine - INFO - MUSA available: False
05/08 15:24:12 - mmengine - INFO - numpy_random_seed: 2147483648
05/08 15:24:12 - mmengine - INFO - GPU 0: NVIDIA GeForce RTX 4090 D
05/08 15:24:12 - mmengine - INFO - CUDA_HOME: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8
05/08 15:24:12 - mmengine - INFO - NVCC: Cuda compilation tools, release 11.8, V11.8.89
05/08 15:24:12 - mmengine - INFO - MSVC: 用于 x64 的 Microsoft (R) C/C++ 优化编译器 19.43.34808 版
05/08 15:24:12 - mmengine - INFO - GCC: n/a
05/08 15:24:12 - mmengine - INFO - PyTorch: 2.0.1
05/08 15:24:12 - mmengine - INFO - PyTorch compiling details: PyTorch built with:
  - C++ Version: 199711
  - MSVC 193431937
  - Intel(R) Math Kernel Library Version 2020.0.2 Product Build 20200624 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v2.7.3 (Git Hash 6dbeffbae1f23cbbeae17adb7b5b13f1f37c080e)
  - OpenMP 2019
  - LAPACK is enabled (usually provided by MKL)
  - CPU capability usage: AVX2
  - CUDA Runtime 11.8
  - 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_90,code=sm_90;-gencode;arch=compute_37,code=compute_37
  - CuDNN 8.7
  - Magma 2.5.4
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.8, CUDNN_VERSION=8.7.0, CXX_COMPILER=C:/cb/pytorch_1000000000000/work/tmp_bin/sccache-cl.exe, CXX_FLAGS=/DWIN32 /D_WINDOWS /GR /EHsc /w /bigobj /FS -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_DISABLE_GPU_ASSERTS=OFF, TORCH_VERSION=2.0.1, 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=OFF, USE_NNPACK=OFF, USE_OPENMP=ON, USE_ROCM=OFF, 

05/08 15:24:12 - mmengine - INFO - TorchVision: 0.15.2
05/08 15:24:12 - mmengine - INFO - OpenCV: 4.11.0
05/08 15:24:12 - mmengine - INFO - MMEngine: 0.10.7
05/08 15:24:12 - mmengine - INFO - MMCV: 2.0.1
05/08 15:24:12 - mmengine - INFO - MMCV Compiler: MSVC 192930148
05/08 15:24:12 - mmengine - INFO - MMCV CUDA Compiler: 11.8
05/08 15:24:12 - mmengine - INFO - MMDeploy: 1.3.1+
05/08 15:24:12 - mmengine - INFO - 

05/08 15:24:12 - mmengine - INFO - **********Backend information**********
05/08 15:24:12 - mmengine - INFO - tensorrt:	None
05/08 15:24:12 - mmengine - INFO - ONNXRuntime:	1.17.3
05/08 15:24:12 - mmengine - INFO - ONNXRuntime-gpu:	1.17.1
05/08 15:24:12 - mmengine - INFO - ONNXRuntime custom ops:	Available
05/08 15:24:12 - mmengine - INFO - pplnn:	None
05/08 15:24:12 - mmengine - INFO - ncnn:	None
05/08 15:24:12 - mmengine - INFO - snpe:	None
05/08 15:24:12 - mmengine - INFO - openvino:	None
05/08 15:24:12 - mmengine - INFO - torchscript:	2.0.1+cu118
05/08 15:24:12 - mmengine - INFO - torchscript custom ops:	NotAvailable
05/08 15:24:12 - mmengine - INFO - rknn-toolkit:	None
05/08 15:24:12 - mmengine - INFO - rknn-toolkit2:	None
05/08 15:24:12 - mmengine - INFO - ascend:	None
05/08 15:24:12 - mmengine - INFO - coreml:	None
05/08 15:24:12 - mmengine - INFO - tvm:	None
05/08 15:24:12 - mmengine - INFO - vacc:	None
05/08 15:24:12 - mmengine - INFO - 

05/08 15:24:12 - mmengine - INFO - **********Codebase information**********
05/08 15:24:12 - mmengine - INFO - mmdet:	3.1.0
05/08 15:24:12 - mmengine - INFO - mmseg:	1.2.1
05/08 15:24:12 - mmengine - INFO - mmpretrain:	None
05/08 15:24:12 - mmengine - INFO - mmocr:	None
05/08 15:24:12 - mmengine - INFO - mmagic:	None
05/08 15:24:12 - mmengine - INFO - mmdet3d:	None
05/08 15:24:12 - mmengine - INFO - mmpose:	None
05/08 15:24:12 - mmengine - INFO - mmrotate:	1.0.0rc1
05/08 15:24:12 - mmengine - INFO - mmaction:	None
05/08 15:24:12 - mmengine - INFO - mmrazor:	None
05/08 15:24:12 - mmengine - INFO - mmyolo:	None

Error traceback