解决ValueError: Unsupported ONNX opset version: 16问题

简介: 解决ValueError: Unsupported ONNX opset version: 16问题

转ONNX模型报错

Traceback (most recent call last):
  File "/root/miniconda3/lib/python3.9/runpy.py", line 197, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/root/miniconda3/lib/python3.9/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/__main__.py", line 39, in <module>
    cli.main()
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 430, in main
    run()
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 284, in run_file
    runpy.run_path(target, run_name="__main__")
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 321, in run_path
    return _run_module_code(code, init_globals, run_name,
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 135, in _run_module_code
    _run_code(code, mod_globals, init_globals,
  File "/root/.vscode-server/extensions/ms-python.python-2022.20.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 124, in _run_code
    exec(code, run_globals)
  File "/cephfs/HZ-AI/usr/hezhiqiang/dou-ai/src/tools/onnx_model.py", line 73, in <module>
    onnx_model.load_weights(model_path='/cephfs/HZ-AI/project/doudizhu/experiments/exp.20221122191411/models/learner_cid_0')
  File "/cephfs/HZ-AI/usr/hezhiqiang/dou-ai/src/tools/onnx_model.py", line 35, in load_weights
    self.export_onnx()
  File "/cephfs/HZ-AI/usr/hezhiqiang/dou-ai/src/tools/onnx_model.py", line 15, in export_onnx
    torch.onnx.export(
  File "/root/miniconda3/lib/python3.9/site-packages/torch/onnx/__init__.py", line 316, in export
    return utils.export(model, args, f, export_params, verbose, training,
  File "/root/miniconda3/lib/python3.9/site-packages/torch/onnx/utils.py", line 107, in export
    _export(model, args, f, export_params, verbose, training, input_names, output_names,
  File "/root/miniconda3/lib/python3.9/site-packages/torch/onnx/utils.py", line 707, in _export
    _set_opset_version(opset_version)
  File "/root/miniconda3/lib/python3.9/site-packages/torch/onnx/symbolic_helper.py", line 849, in _set_opset_version
    raise ValueError("Unsupported ONNX opset version: " + str(opset_version))
ValueError: Unsupported ONNX opset version: 16

修复步骤

  1. 安装PyTorch1.12.1
pip install torch==1.12.1 -ihttps://pypi.tuna.tsinghua.edu.cn/simple
  1. 卸载之前的horovod
pip uninstall horovod
  1. 更新cmake
pip install cmake --upgrade -ihttps://pypi.tuna.tsinghua.edu.cn/simple
  1. 安装horovod
HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_MXNET=1 HOROVOD_WITH_PYTORCH=1 HOROVOD_WITHOUT_GLOO=1 HOROVOD_GPU_OPERATIONS=NCCL HOROVOD_CMAKE=/root/miniconda3/bin/cmake /root/miniconda3/bin/pip install --verbose --no-cache-dir horovod

安装horovod的时候需要GPU。

相关文章
Jetson学习笔记(二):TensorRT 查看模型的输入输出
这篇博客介绍了如何使用TensorRT查看模型的输入输出,并通过代码示例展示了如何获取和验证模型的输入输出信息。
529 5
|
机器学习/深度学习 算法 测试技术
低照度增强算法(图像增强+目标检测+代码)
低照度增强算法(图像增强+目标检测+代码)
|
负载均衡 Ubuntu 应用服务中间件
|
11月前
|
存储 人工智能 算法
【AI系统】计算图的优化策略
本文深入探讨了计算图的优化策略,包括算子替换、数据类型转换、存储优化等,旨在提升模型性能和资源利用效率。特别介绍了Flash Attention算法,通过分块计算和重算策略优化Transformer模型的注意力机制,显著减少了内存访问次数,提升了计算效率。此外,文章还讨论了内存优化技术,如Inplace operation和Memory sharing,进一步减少内存消耗,提高计算性能。
574 34
【AI系统】计算图的优化策略
|
XML 机器学习/深度学习 数据格式
YOLOv8训练自己的数据集+常用传参说明
YOLOv8训练自己的数据集+常用传参说明
21080 3
|
编解码 计算机视觉 异构计算
【CV大模型SAM(Segment-Anything)】如何一键分割图片中所有对象?并对不同分割对象进行保存?
【CV大模型SAM(Segment-Anything)】如何一键分割图片中所有对象?并对不同分割对象进行保存?
|
Ubuntu 计算机视觉 C++
Ubuntu 20.04 编译 Opencv 4.11,详细步骤(带图)及报错解决,我的踩坑之旅~
Ubuntu 20.04 编译 Opencv 4.11,详细步骤(带图)及报错解决,我的踩坑之旅~
9487 0
|
机器学习/深度学习 并行计算 PyTorch
【已解决】RuntimeError: CUDA error: device-side assert triggeredCUDA kernel errors might be asynchronous
【已解决】RuntimeError: CUDA error: device-side assert triggeredCUDA kernel errors might be asynchronous
10252 2
|
机器学习/深度学习 编解码 人工智能
Transformer 和扩散模型的生成式 AI 实用指南(预览版)
Transformer 和扩散模型的生成式 AI 实用指南(预览版)
855 1
Transformer 和扩散模型的生成式 AI 实用指南(预览版)
|
存储 自然语言处理 Shell
Transformers 4.37 中文文档(七十六)(1)
Transformers 4.37 中文文档(七十六)
244 0
下一篇
oss云网关配置