PyTorch模型定义

pytorch模型定义主要包括两个部分:各部分初始化init和数据流向定义forword

1.Sequential

当模型的前向计算为简单串联各个层的计算时， Sequential 类可以通过更加简单的方式定义模型.

class MySequential(nn.Module):
    from collections import OrderedDict
    def __init__(self, *args):
        super(MySequential, self).__init__()
        if len(args) == 1 and isinstance(args[0], OrderedDict): # 如果传入的是一个OrderedDict
            for key, module in args[0].items():
                self.add_module(key, module)  
                # add_module方法会将module添加进self._modules(一个OrderedDict)
        else:  # 传入的是一些Module
            for idx, module in enumerate(args):
                self.add_module(str(idx), module)
    def forward(self, input):
        # self._modules返回一个 OrderedDict，保证会按照成员添加时的顺序遍历成
        for module in self._modules.values():
            input = module(input)
        return input

只需将模型的层按序排列即可定义模型
直接排列:

import torch.nn as nn
net = nn.Sequential(
      nn.Linear(784, 256),
      nn.ReLU(),
      nn.Linear(256, 10), 
      )
print(net)

Sequential(
  (0): Linear(in_features=784, out_features=256, bias=True)
  (1): ReLU()
  (2): Linear(in_features=256, out_features=10, bias=True)
)

使用OrderedDict：

import collections
import torch.nn as nn
net2 = nn.Sequential(collections.OrderedDict([
          ('fc1', nn.Linear(784, 256)),
          ('relu1', nn.ReLU()),
          ('fc2', nn.Linear(256, 10))
          ]))
print(net2)

Sequential(
  (fc1): Linear(in_features=784, out_features=256, bias=True)
  (relu1): ReLU()
  (fc2): Linear(in_features=256, out_features=10, bias=True)
)

使用Sequential定义的模型不需要再写forward，因为顺序已经定义好了。但使用Sequential也会使得模型定义丧失灵活性，比如需要在模型中间加入一个外部输入时就不适合用Sequential的方式实现。使用时需根据实际需求加以选择。

2.ModuleList

ModuleList 接收一个子模块的列表作为输入，然后也可以类似List那样进行append和extend操作.

net = nn.ModuleList([nn.Linear(784, 256), nn.ReLU()])
net.append(nn.Linear(256, 10)) # # 类似List的append操作
print(net[-1])  # 类似List的索引访问
print(net)

Linear(in_features=256, out_features=10, bias=True)
ModuleList(
  (0): Linear(in_features=784, out_features=256, bias=True)
  (1): ReLU()
  (2): Linear(in_features=256, out_features=10, bias=True)
)

nn.ModuleList 并没有定义一个网络，它只是将不同的模块储存在一起。ModuleList中元素的先后顺序并不代表其在网络中的真实位置顺序，需要经过forward函数指定各个层的先后顺序后才算完成了模型的定义。具体实现时用for循环即可完成：

class model(nn.Module):
  def __init__(self, ...):
    super().__init__()
    self.modulelist = ...
    ...
    
  def forward(self, x):
    for layer in self.modulelist:
      x = layer(x)
    return x

3.ModuleDict

ModuleDict和ModuleList的作用类似，只是ModuleDict能够更方便地为神经网络的层添加名称。

   net = nn.ModuleDict({
   'linear': nn.Linear(784, 256),
   'act': nn.ReLU(),
})
net['output'] = nn.Linear(256, 10) # 添加
print(net['linear']) # 访问
print(net.output)
print(net)

Linear(in_features=784, out_features=256, bias=True)
Linear(in_features=256, out_features=10, bias=True)
ModuleDict(
  (act): ReLU()
  (linear): Linear(in_features=784, out_features=256, bias=True)
  (output): Linear(in_features=256, out_features=10, bias=True)
)

三种方法比较:
Sequential适用于快速验证结果，因为已经明确了要用哪些层，直接写一下就好了，不需要同时写__init__和forward；
ModuleList和ModuleDict在某个完全相同的层需要重复出现多次时，非常方便实现，可以”一行顶多行“；

利用模型块快速搭建网络

对于大部分模型结构,考虑到每一层有其输入和输出，若干层串联成的”模块“也有其输入和输出，如果我们能将这些重复出现的层定义为一个”模块“，每次只需要向网络中添加对应的模块来构建模型，这样将会极大便利模型构建的过程。

U-Net介绍

U-Net模型通过残差连接结构解决了模型学习中的退化问题，使得神经网络的深度能够不断扩展。
组成U-Net的模型块主要有如下几个部分：
1）每个子块内部的两次卷积（Double Convolution）

2）左侧模型块之间的下采样连接，即最大池化（Max pooling）

3）右侧模型块之间的上采样连接（Up sampling）

4）输出层的处理

除模型块外，还有模型块之间的横向连接，输入和U-Net底部的连接等计算，这些单独的操作可以通过forward函数来实现。
U-Net模型块实现:

import torch
import torch.nn as nn
import torch.nn.functional as F

class DoubleConv(nn.Module):
    """(convolution => [BN] => ReLU) * 2"""

    def __init__(self, in_channels, out_channels, mid_channels=None):
        super().__init__()
        if not mid_channels:
            mid_channels = out_channels
        self.double_conv = nn.Sequential(
            nn.Conv2d(in_channels, mid_channels, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(mid_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(mid_channels, out_channels, kernel_size=3, padding=1, bias=False),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True)
        )

    def forward(self, x):
        return self.double_conv(x)

class Down(nn.Module):
"""Downscaling with maxpool then double conv"""

def __init__(self, in_channels, out_channels):
    super().__init__()
    self.maxpool_conv = nn.Sequential(
        nn.MaxPool2d(2),
        DoubleConv(in_channels, out_channels)
    )

def forward(self, x):
    return self.maxpool_conv(x)

class Up(nn.Module):
    """Upscaling then double conv"""

    def __init__(self, in_channels, out_channels, bilinear=False):
        super().__init__()

        # if bilinear, use the normal convolutions to reduce the number of channels
        if bilinear:
            self.up = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=True)
            self.conv = DoubleConv(in_channels, out_channels, in_channels // 2)
        else:
            self.up = nn.ConvTranspose2d(in_channels, in_channels // 2, kernel_size=2, stride=2)
            self.conv = DoubleConv(in_channels, out_channels)

    def forward(self, x1, x2):
        x1 = self.up(x1)
        # input is CHW
        diffY = x2.size()[2] - x1.size()[2]
        diffX = x2.size()[3] - x1.size()[3]

        x1 = F.pad(x1, [diffX // 2, diffX - diffX // 2,
                        diffY // 2, diffY - diffY // 2])
        # if you have padding issues, see
        # https://github.com/HaiyongJiang/U-Net-Pytorch-Unstructured-Buggy/commit/0e854509c2cea854e247a9c615f175f76fbb2e3a
        # https://github.com/xiaopeng-liao/Pytorch-UNet/commit/8ebac70e633bac59fc22bb5195e513d5832fb3bd
        x = torch.cat([x2, x1], dim=1)
        return self.conv(x)

class OutConv(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(OutConv, self).__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=1)

    def forward(self, x):
        return self.conv(x)

利用模型块组装U-Net

使用写好的模型块，可以非常方便地组装U-Net模型。

class UNet(nn.Module):
  def __init__(self, n_channels, n_classes, bilinear=False):
      super(UNet, self).__init__()
      self.n_channels = n_channels
      self.n_classes = n_classes
      self.bilinear = bilinear

      self.inc = DoubleConv(n_channels, 64)
      self.down1 = Down(64, 128)
      self.down2 = Down(128, 256)
      self.down3 = Down(256, 512)
      factor = 2 if bilinear else 1
      self.down4 = Down(512, 1024 // factor)
      self.up1 = Up(1024, 512 // factor, bilinear)
      self.up2 = Up(512, 256 // factor, bilinear)
      self.up3 = Up(256, 128 // factor, bilinear)
      self.up4 = Up(128, 64, bilinear)
      self.outc = OutConv(64, n_classes)

  def forward(self, x):
      x1 = self.inc(x)
      x2 = self.down1(x1)
      x3 = self.down2(x2)
      x4 = self.down3(x3)
      x5 = self.down4(x4)
      x = self.up1(x5, x4)
      x = self.up2(x, x3)
      x = self.up3(x, x2)
      x = self.up4(x, x1)
      logits = self.outc(x)
      return logits

参考链接:https://datawhalechina.github.io/thorough-pytorch/%E7%AC%AC%E4%BA%94%E7%AB%A0/5.2%20%E5%88%A9%E7%94%A8%E6%A8%A1%E5%9E%8B%E5%9D%97%E5%BF%AB%E9%80%9F%E6%90%AD%E5%BB%BA%E5%A4%8D%E6%9D%82%E7%BD%91%E7%BB%9C.html

原文链接：https://blog.csdn.net/Nature_park/article/details/126456833