mxnet转onnx

这个还有onnx推理代码和报错解决

Insightface中ArcFace MxNet2ONNX踩坑 - 知乎

转换代码：

import numpy as np
import mxnet as mx
from mxnet.contrib import onnx as onnx_mxnet
 
sym = './model-symbol.json'
params = './model-0000.params'
input_shape = (1, 3, 112, 112)
onnx_file = './mobilefacenet.onnx'
 
converted_model_path = onnx_mxnet.export_model(sym, params, [input_shape], np.float32, onnx_file)

onnx1.9 报错：

spatial for operator BatchNormalization

onnx.onnx_cpp2py_export.checker.ValidationError: Unrecognized attribute: spatial for operator BatchNormalization

==> Context: Bad node spec: input: "conv_1_conv2d" input: "conv_1_batchnorm_gamma" input: "conv_1_batchnorm_beta" input: "conv_1_batchnorm_moving_mean" input: "conv_1_batchnorm_moving_var" output: "conv_1_batchnorm" name: "conv_1_batchnorm" op_type: "BatchNormalization" attribute { name: "epsilon" f: 0.001 type: FLOAT } attribute { name: "momentum" f: 0.9 type: FLOAT } attribute { name: "spatial" i: 0 type: INT }
 

解决方法：

请看这篇文章：https://zhuanlan.zhihu.com/p/165294876?utm_source=ZHShareTargetIDMore

用root权限编辑文件 /usr/local/lib/python3.6/dist-packages/mxnet/contrib/onnx/mx2onnx/_op_translations.py

把 359行的 spatial=0 屏蔽掉即可。
测试确实可以转：

 原文链接：https://blog.csdn.net/u012505617/article/details/110530821