Description
🐛 Describe the bug
try to quantize a model like this link
(only different in model structures and datasets)
then export the quantized model to onnx by torch.onnx.export (original model is able to output), and get
File "d:\my_project\train_quantized.py", line 798, in <module>
onnx_program = torch.onnx.export(model, torch_input, "my_quantized.onnx")
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\__init__.py", line 375, in export
export(
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\utils.py", line 502, in export
_export(
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\utils.py", line 1564, in _export
graph, params_dict, torch_out = _model_to_graph(
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\utils.py", line 1117, in _model_to_graph
graph = _optimize_graph(
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\utils.py", line 639, in _optimize_graph
graph = _C._jit_pass_onnx(graph, operator_export_type)
File "D:\anaconda3\envs\hm\lib\site-packages\torch\onnx\utils.py", line 1848, in _run_symbolic_function
raise errors.UnsupportedOperatorError(
torch.onnx.errors.UnsupportedOperatorError: ONNX export failed on an operator with unrecognized namespace quantized_decomposed::quantize_per_tensor. If you are trying to export a custom operator, make sure you registered it with the right domain and version.
Versions
Collecting environment information...
PyTorch version: 2.5.1
Is debug build: False
CUDA used to build PyTorch: 11.8
ROCM used to build PyTorch: N/A
OS: Microsoft Windows 11 专业版 (10.0.22631 64 位)
GCC version: Could not collect
Clang version: Could not collect
CMake version: Could not collect
Libc version: N/A
Python version: 3.10.14 | packaged by conda-forge | (main, Mar 20 2024, 12:40:08) [MSC v.1938 64 bit (AMD64)] (64-bit runtime)
Python platform: Windows-10-10.0.22631-SP0
Is CUDA available: True
CUDA runtime version: 11.8.89
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4090 D
Nvidia driver version: 560.94
cuDNN version: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8\bin\cudnn_ops_train64_8.dll
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
CPU:
Name: Intel(R) Core(TM) i9-14900KF
Manufacturer: GenuineIntel
Family: 207
Architecture: 9
ProcessorType: 3
DeviceID: CPU0
CurrentClockSpeed: 3200
MaxClockSpeed: 3200
L2CacheSize: 32768
L2CacheSpeed: None
Revision: None
Versions of relevant libraries:
[pip3] efficientnet_pytorch==0.7.1
[pip3] flake8==7.1.0
[pip3] mypy-extensions==1.0.0
[pip3] numpy==2.1.3
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] >
[pip3] onnx-tool==0.9.0
[pip3] >
[pip3] >
[pip3] >
[pip3] optree==0.12.1
[pip3] pytorch-lightning==2.4.0
[pip3] segmentation-models-pytorch==0.3.4
[pip3] torch==2.5.1
[pip3] torch-pruning==1.5.1
[pip3] torch-tb-profiler==0.4.3
[pip3] torch_tensorrt==2.5.0
[pip3] torchaudio==2.5.1
[pip3] torchmetrics==1.4.2
[pip3] torchvision==0.20.1
[conda] blas 1.0 mkl defaults
[conda] cuda-cudart 11.8.89 0 nvidia
[conda] cuda-cudart-dev 11.8.89 0 nvidia
[conda] cuda-cupti 11.8.87 0 nvidia
[conda] cuda-libraries 11.8.0 0 nvidia
[conda] cuda-libraries-dev 11.8.0 0 nvidia
[conda] cuda-nvrtc 11.8.89 0 nvidia
[conda] cuda-nvrtc-dev 11.8.89 0 nvidia
[conda] cuda-nvtx 11.8.86 0 nvidia
[conda] cuda-opencl 12.5.39 he0c23c2_1 conda-forge
[conda] cuda-opencl-dev 12.5.39 he0c23c2_1 conda-forge
[conda] cuda-runtime 11.8.0 0 nvidia
[conda] efficientnet-pytorch 0.7.1 pypi_0 pypi
[conda] libcublas 11.11.3.6 0 nvidia
[conda] libcublas-dev 11.11.3.6 0 nvidia
[conda] libcufft 10.9.0.58 0 nvidia
[conda] libcufft-dev 10.9.0.58 0 nvidia
[conda] libcurand 10.3.6.82 he0c23c2_0 conda-forge
[conda] libcurand-dev 10.3.6.82 he0c23c2_0 conda-forge
[conda] libcusolver 11.4.1.48 0 nvidia
[conda] libcusolver-dev 11.4.1.48 0 nvidia
[conda] libcusparse 11.7.5.86 0 nvidia
[conda] libcusparse-dev 11.7.5.86 0 nvidia
[conda] libnvjitlink 12.4.127 0 nvidia
[conda] libnvjitlink-dev 12.4.127 0 nvidia
[conda] mkl 2021.4.0 pypi_0 pypi
[conda] mkl-service 2.4.0 py310h2bbff1b_1 defaults
[conda] mkl_fft 1.3.11 py310h827c3e9_0 defaults
[conda] mkl_random 1.2.8 py310hc64d2fc_0 defaults
[conda] numpy 2.1.3 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu12 12.6.77 pypi_0 pypi
[conda] optree 0.12.1 pypi_0 pypi
[conda] pytorch 2.5.1 py3.10_cuda11.8_cudnn9_0 pytorch
[conda] pytorch-cuda 11.8 h24eeafa_6 pytorch
[conda] pytorch-lightning 2.4.0 pypi_0 pypi
[conda] pytorch-mutex 1.0 cuda pytorch
[conda] segmentation-models-pytorch 0.3.4 pypi_0 pypi
[conda] torch-pruning 1.5.1 pypi_0 pypi
[conda] torch-tb-profiler 0.4.3 pypi_0 pypi
[conda] torch-tensorrt 2.5.0 pypi_0 pypi
[conda] torchaudio 2.5.1 pypi_0 pypi
[conda] torchmetrics 1.4.2 pypi_0 pypi
[conda] torchvision 0.20.1 pypi_0 pypi