Pytorch的nn.Module類中定義的實例方法apply()

參考文檔Module — PyTorch 1.7.0 documentation

 @torch.no_grad()
 def init_weights(m):
     print(m)
     if type(m) == nn.Linear:
         m.weight.fill_(1.0)
         print(m.weight)
 net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
 net.apply(init_weights)

net類及其子類都會調用 init_weights() 方法