基于pytorch框架实现的网络中自定义了如下函数：

def weight_init(module):
    for n,m in module.named_children():
        if isinstance(m,nn.Conv2d):
            nn.init.kaiming_normal_(m.weight,mode='fan_in',nonlinearity='relu')
            if m.bias is not None:
                nn.init.zeros_(m.bias)

Paddle框架下的paddle.nn.initializer也提供了nn.initializer.KaimingNormal()、nn.initializer.nn.initializer.Constant()函数，但用法与别的框架不一样。参考Paddle文档后对函数做出修改：

def weight_init(Layer):
    for n, m in Layer.named_children():
        if isinstance(m, nn.Conv2D):
            m.weight = nn.Linear(weight_attr = nn.initializer.KaimingNormal())
            if m.bias is not None:
                m.bias = nn.Linear(weight_attr = nn.initializer.Constant(value = 0.))

但此时Linear层缺少in_channels、out_channels，即对应的m_weight、m_bias的维度信息。

求教，如何去求解迭代器返回的m中weight、m_bias的维度信息，或者有其它的写法来实现函数功能