我创建了一个称为CustomFunc的自定义功能,说明这里下面:https://www.cntk.ai/pythondocs/extend.html如何编写自定义函数CNTK
如果我使用它的文章的建议,它的工作原理:
model = cntk.user_function(CustomFunc(prev_node))
这个作品很好,模型运行没有任何问题。我的问题是,我想在cntk.layers.Sequential调用中使用此函数,并在cntk.layers.Recurrence调用中使用此函数。要做到这一点,我需要以另一种方式构建函数的组合,然后将其放入Sequential或Recurrence调用中。现在我使用一些占位符,即我做的是:
customFunToUse = cntk.user_function(CustomFunc(cntk.placeholder(), otherInputs))
model = cntk.layers.Sequential([cntk.layers.Dense(100),
customFunToUse,
cntk.layers.Recurrence(
customFunToUse >> cntk.layers.LSTM(100))])
但是,这并不工作,并提出了各种错误:有时它是一个段错误,在其他类似型号是
"ValueError: Cannot create an NDArrayView using a view shape '[? x 10]' that has unknown dimensions for any of its axes."
而不是
其他时间是
Evaluate: All nodes inside a recurrent loop must have a layout that is identical; mismatch found for nodes ...
还要注意的是我的自定义功能不改变输入尺寸:给予paramters的任何金额,它会返回相同的数量和类型。该代码是这样的:
class CustomFun(UserFunction):
def __init__(self, *args, otherStuff, name='CustomFun'):
super(CustomFun, self).__init__(list(args), name=name)
self.otherStuff = otherStuff
def forward(self, arguments, outputs=None, keep_for_backward=None, device=None, as_numpy=True):
return None,[x/2 for x in arguments]
def backward(self, state, root_gradients, variables=None, as_numpy=True):
#it's not important right now, just a test...
return root_gradient
def infer_outputs(self):
#shape, type and dynamic axes of inputs are not changed by this function
outputVar = [output_variable(self.inputs[idx].shape, self.inputs[idx].dtype,
self.inputs[idx].dynamic_axes, name='out_quantLayer') for idx in range(len(self.inputs))]
return outputVar
def serialize(self):
return {'otherStuff': self.otherStuff}
@staticmethod
def deserialize(inputs, name, state):
return CustomFun(inputs, otherStuff=state['otherStuff'], name=name)