使用Sequential和Module两种方式搭建模型

使用Sequential方式进行建模 

import numpy as np
import torch
from torch import nn

# 定义一个 Sequential 模型
net1 = nn.Sequential(
    nn.Linear(30, 40),
    nn.ReLU(),
    nn.Linear(40, 50),
    nn.ReLU(),
    nn.Linear(50, 10)
)

for layer in net1:
    print(layer)
    if isinstance(layer, nn.Linear): # 判断是否是线性层
        param_shape = layer.weight.shape
        layer.weight.data = torch.from_numpy(np.random.normal(0, 0.5, size=param_shape)) 
        # 定义为均值为 0，方差为 0.5 的正态分布

输出结果如下： 

Linear(in_features=30, out_features=40, bias=True)
ReLU()
Linear(in_features=40, out_features=50, bias=True)
ReLU()
Linear(in_features=50, out_features=10, bias=True)

使用Module方式进行建模 

对于 Module 的参数初始化，其实也非常简单，如果想对其中的某层进行初始化，可以直接像 Sequential 一样对其 Tensor 进行重新定义，其唯一不同的地方在于，如果要用循环的方式访问，需要介绍两个属性，children 和 modules，下面我们举例来说明

class sim_net(nn.Module):
    def __init__(self):
        super(sim_net, self).__init__()
        self.l1 = nn.Sequential(
            nn.Linear(30, 40),
            nn.ReLU()
        )
        
        self.l1[0].weight.data = torch.randn(40, 30) # 直接对某一层初始化
        
        self.l2 = nn.Sequential(
            nn.Linear(40, 50),
            nn.ReLU()
        )
        
        self.l3 = nn.Sequential(
            nn.Linear(50, 10),
            nn.ReLU()
        )
    
    def forward(self, x):
        x = self.l1(x)
        x =self.l2(x)
        x = self.l3(x)
        return x

i = 0
for layer in net2.modules():
    print(i)
    i = i + 1
    if isinstance(layer, nn.Linear):
        param_shape = layer.weight.shape
        layer.weight.data = torch.from_numpy(np.random.normal(0, 0.5, size=param_shape)) 

i = 0
for layer in net2.children():
    print(i)
    i = i + 1
    if isinstance(layer[0], nn.Linear):
        param_shape = layer[0].weight.shape
        layer[0].weight.data = torch.from_numpy(np.random.normal(0, 0.5, size=param_shape))

遍历module属性的时候，输出结果如下：

0
1
2
3
4
5
6
7
8
9

遍历children属性的时候，输出结果如下：

0
1
2

比较上述两种输出结果，不难发现，使用module方式进行建模的时候，我们是从顶层到底层对模型进行遍历，而采用chidren方式则只能对模型的Sequetial进行遍历。下面我们就具体来看看两种遍历方式的不同：

# 访问 modules
for i in net2.modules():
    print(i)

sim_net(
  (l1): Sequential(
    (0): Linear(in_features=30, out_features=40)
    (1): ReLU()
  )
  (l2): Sequential(
    (0): Linear(in_features=40, out_features=50)
    (1): ReLU()
  )
  (l3): Sequential(
    (0): Linear(in_features=50, out_features=10)
    (1): ReLU()
  )
)
Sequential(
  (0): Linear(in_features=30, out_features=40)
  (1): ReLU()
)
Linear(in_features=30, out_features=40)
ReLU()
Sequential(
  (0): Linear(in_features=40, out_features=50)
  (1): ReLU()
)
Linear(in_features=40, out_features=50)
ReLU()
Sequential(
  (0): Linear(in_features=50, out_features=10)
  (1): ReLU()
)
Linear(in_features=50, out_features=10)
ReLU()

# 访问 children
for i in net2.children():
    print(i)

Sequential(
  (0): Linear(in_features=30, out_features=40, bias=True)
  (1): ReLU()
)
Sequential(
  (0): Linear(in_features=40, out_features=50, bias=True)
  (1): ReLU()
)
Sequential(
  (0): Linear(in_features=50, out_features=10, bias=True)
  (1): ReLU()
)