PROSAGA码农传奇-深度学习-Mxnet版本1.0.0胶子KeyError：'shape'

Mxnet版本1.0.0胶子KeyError：'shape'

作者: 与之
发布时间: 2025-01-16 09:38:57 (1月前)
转自：

<div class =“post-text”itemprop =“text”>
  <P>
    要检查权重，首先需要按如下方式初始化它们：

<code>
      
net.collect_params().initialize(mx.init.Xavier(), ctx=mx.cpu())

</code>
  </p>
  <P>
    在您的情况下，因为您没有在Net的构造函数中指定图层的输入大小，所以此时无法确定参数的形状。所以，如果你访问
     <code>
      net.weight.data()
    </code>
     现在，将引发一个例外：
  </p>
  <P>
     <code>
      
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/dist-packages/mxnet-0.12.1-py2.7.egg/mxnet/gluon/parameter.py", line 389, in data
    return self._check_and_get(self._data, ctx)
  File "/usr/local/lib/python2.7/dist-packages/mxnet-0.12.1-py2.7.egg/mxnet/gluon/parameter.py", line 189, in _check_and_get
    "nested child Blocks"%(self.name))
RuntimeError: Parameter dense0_weight has not been initialized. Note that you should initialize parameters and create Trainer with Block.collect_params() instead of Block.params because the later does not include Parameters of nested child Blocks

</code>
  </p>
  <P>
    您可以按如下方式初始化权重：

<code>
      
net.collect_params().initialize(mxnet.init.Xavier(), ctx=mxnet.cpu())
print (net.weight.data())

</code>
  </p>
</DIV>