医学分割模型 peddleseg

医疗标注平台使用教程

注意

仅可使用cuda10.0版本

Python 3.9.10 (main, Mar  2 2025, 20:55:00) [MSC v.1942 64 bit (AMD64)] on win32
>>> 
[Qt] void __cdecl qSlicerMarkupsPlaceWidget::setPlaceModePersistency(bool)  failed: interactionNode is invalid
[Qt] qSlicerMarkupsPlaceWidget::setDeleteAllMarkupsOptionVisible method is deprecated, please use setDeleteAllControlPointsOptionVisible instead
[Qt] qMRMLSegmentEditorWidget::setMasterVolumeNodeSelectorVisible is deprecated, use setSourceVolumeNodeSelectorVisible method instead.
Traceback (most recent call last):
  File "G:/ovcfimage/PaddleSeg/EISeg/med3d/EISegMed3D/EISegMed3D.py", line 1039, in loadModelClicked
    self.inference_predictor = predictor.BasePredictor(
  File "G:\ovcfimage\PaddleSeg\EISeg\med3d\EISegMed3D\inference\predictor.py", line 65, in __init__
    self.normalization = BatchImageNormalize3D(
  File "G:\ovcfimage\PaddleSeg\EISeg\med3d\EISegMed3D\inference\ops.py", line 50, in __init__
    self.mean = paddle.to_tensor(
  File "E:\slicer\Slicer 5.8.1\lib\Python\Lib\site-packages\paddle\base\dygraph\math_op_patch.py", line 92, in astype
    return _C_ops.cast(self, dtype)
RuntimeError: (PreconditionNotMet) The third-party dynamic library (cudnn64_8.dll) that Paddle depends on is not configured correctly. (error code is 126)
  Suggestions:
  1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
  2. Configure third-party dynamic library environment variables as follows:
  - Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...`
  - Windows: set PATH by `set PATH=XXX; (at ..\paddle\phi\backends\dynload\dynamic_loader.cc:312)

[FD] W0317 17:11:15.494228  3860 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 6.1, Driver API Version: 12.6, Runtime API Version: 11.8
[FD] W0317 17:11:15.495229  3860 dynamic_loader.cc:285] Note: [Recommend] copy cudnn into CUDA installation directory. 
[FD]  For instance, download cudnn-10.0-windows10-x64-v7.6.5.32.zip from NVIDIA's official website, 
[FD] then, unzip it and copy it into C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0
[FD] You should do this according to your CUDA installation directory and CUDNN version.

需要用到的cudnn版本

https://developer.download.nvidia.com/compute/redist/cudnn/v7.6.5/cudnn-10.0-windows10-x64-v7.6.5.32.zip

建议使用monai