在Python中如何在PyTorch中初始化权重？

如何在PyTorch中的网络中初始化权重和偏差(例如,使用He或Xavier初始化)？ 单层

要初始化单个图层的权重,请使用 torch.nn.init 中的函数.例如：

conv1 = torch.nn.Conv2d(...)
torch.nn.init.xavier_uniform(conv1.weight)

或者,您可以通过写入conv1.weight.data(这是一个 torch.Tensor )来修改参数.例：

conv1.weight.data.fill_(0.01)

这同样适用于偏见：

conv1.bias.data.fill_(0.01)

nn.Sequential或custom nn.Module

将初始化函数传递给 torch.nn.Module.apply .它将递归地初始化整个nn.Module中的权重.

apply( fn ): Applies fn recursively to every submodule (as returned by .children() ) as well as self. Typical use includes initializing the parameters of a model (see also torch-nn-init).

例：

def init_weights(m):
    if type(m) == nn.Linear:
        torch.nn.init.xavier_uniform(m.weight)
        m.bias.data.fill_(0.01)

net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)

查看更多关于在Python中如何在PyTorch中初始化权重？的详细内容...