模型导出错误：'EncoderDecoder' object has no attribute 'forward_dummy'

[shuaizzZ]

1、先是报如下错误： Traceback (most recent call last): File "/mnt/bn/shuaizzz/codes/mmsegmentation/tools/deployment/pytorch2torchscript_defect.py", line 201, in segmentor = build_segmentor( File "/mnt/bn/shuaizzz/codes/mmsegmentation/mmseg/models/builder.py", line 49, in build_segmentor assert cfg.get('test_cfg') is None or test_cfg is None,
AssertionError: test_cfg specified in both outer field and model field

我通过修改pytorch2torchscript.py文件的segmentor定义为test_cfg=None解决了这个报错。

2、然后又报错： KeyError: 'SegDataPreProcessor is not in the model registry. Please check whether the value of SegDataPreProcessor is correct or it was registered as expected. More details can be found at https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html#import-the-custom-module'

我通过修改pytorch2torchscript.py文件，添加如下两行，解决了上述报错： from mmseg.utils import register_all_modules register_all_modules()

3、随后又报错： Traceback (most recent call last): File "/mnt/bn/shuaizzz/codes/mmsegmentation/tools/deployment/pytorch2torchscript_defect.py", line 213, in pytorch2libtorch( File "/mnt/bn/shuaizzz/codes/mmsegmentation/tools/deployment/pytorch2torchscript_defect.py", line 125, in pytorch2libtorch model.forward = model.forward_dummy File "/root/miniconda3/envs/mmseg/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1207, in getattr raise AttributeError("'{}' object has no attribute '{}'".format( AttributeError: 'EncoderDecoder' object has no attribute 'forward_dummy'

请问该如何解决？

[mm-assistant[bot]]

We recommend using English or English & Chinese for issues so that we could have broader discussion.

[sammilei]

Got the same issue with trained segmenter tiny .pth trained on mmcv=1.6.0 and mmseg==0.24.1 to .torchscript on mmcv=2.0.0! It also failed in mmcv=1.6.0 but I forgot what that issue it was.

Traceback (most recent call last):
  File "tools/deployment/pytorch2torchscript.py", line 180, in <module>
    pytorch2libtorch(
  File "tools/deployment/pytorch2torchscript.py", line 117, in pytorch2libtorch
    model.forward = model.forward_dummy
  File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1185, in __getattr__
    raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'EncoderDecoder' object has no attribute 'forward_dummy'

My system

sys.platform: linux
Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0]
CUDA available: True
numpy_random_seed: 2147483648
GPU 0,1,2,3,4,5,6,7: Tesla V100-SXM2-16GB
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 11.3, V11.3.109
GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
PyTorch: 1.11.0
PyTorch compiling details: PyTorch built with:
  - GCC 7.3
  - C++ Version: 201402
  - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: AVX2
  - 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 2.5.2
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, 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=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, 

TorchVision: 0.12.0
OpenCV: 4.7.0
MMEngine: 0.7.2
MMSegmentation: 1.0.0+b600f7c

[successAI]

I also met the same problem. Have you solved it？

[ZiAn-Su]

delete 'model.forward = model.forward_dummy' in pytorch2torchscript.py

[nothingws]

me too

[nothingws]

We recommend using English or English & Chinese for issues so that we could have broader discussion. May I know if you have resolved the issue?

[geoeo]

This worked for me.