深度学习面试题31：池化操作的梯度

目录

　　平均池化

　　最大值池化

　　参考资料

---

平均池化

 

x和卷积核K做卷积运算，得到σ，对σ的做平均池化，得到S，然后S与GT计算MSE损失。

  

对应代码

# -*- coding: utf-8 -*- 
import torch
import torch.nn as nn
import torch.nn.functional as F


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 1, 2, bias=False)  # 默认padding=0 即valid卷积

    def forward(self, x):
        # Max pooling over a (2, 2) window
        x = self.conv1(x)
        x = F.avg_pool2d(x, 2)
        # x = F.max_pool2d(x, 2)

        return x


if __name__ == '__main__':
    net = Net()
    print("网络结构为：")
    print(net)
    print()
    weight1 = torch.tensor([3., 2., 1., 5.])
    weight1 = weight1.view(1, 1, 2, 2)
    net.conv1._parameters['weight'].data = weight1  # 自定义卷积核

    input = torch.tensor([[1., 2., 3.],  # 自定义输入
                          [4., 5., 6.],
                          [7., 8., 9.]])
    input = input.view(1, 1, 3, 3)
    output = net(input)
    print("前向传播输出：")
    print(output)
    print()

    # Loss Function
    target = torch.tensor(60.)
    criterion = nn.MSELoss()
    loss = criterion(output, target)
    print("MSE loss：", loss)
    print()

    # Backprop
    net.zero_grad()  # zeroes the gradient buffers of all parameters
    loss.backward()
    print("卷积核的梯度：")
    print(net.conv1.weight.grad)
    print()

    use_module = True
    if not use_module:
        # Update the weights     weight = weight - learning_rate * gradient
        learning_rate = 0.01
        for f in net.parameters():
            f.data.sub_(f.grad.data * learning_rate)
        print("手动更新")
        print(list(net.parameters()))
        """
        tensor([[[[2.5200, 1.3600],
                  [0.0400, 3.8800]]]], requires_grad=True)]
        """

    else:
        # However, as you use neural networks, you want to use various different update rules such as SGD,
        # Nesterov-SGD, Adam, RMSProp, etc. To enable this, we built a small package: torch.optim that
        # implements all these methods. Using it is very simple:
        import torch.optim as optim

        # create your optimizer
        optimizer = optim.SGD(net.parameters(), lr=0.01)

        # in your training loop:
        optimizer.zero_grad()  # zero the gradient buffers
        output = net(input)
        loss = criterion(output, target)
        loss.backward()
        optimizer.step()  # Does the update
        print("optim更新")
        print(list(net.parameters()))
        """
        tensor([[[[2.5200, 1.3600],
                  [0.0400, 3.8800]]]], requires_grad=True)]
        """

 返回目录

最大值池化

 

对应代码

# -*- coding: utf-8 -*- 
import torch
import torch.nn as nn
import torch.nn.functional as F


class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 1, 2, bias=False)  # 默认padding=0 即valid卷积

    def forward(self, x):
        # Max pooling over a (2, 2) window
        x = self.conv1(x)
        # x = F.avg_pool2d(x, 2)
        x = F.max_pool2d(x, 2)

        return x


if __name__ == '__main__':
    net = Net()
    print("网络结构为：")
    print(net)
    print()
    weight1 = torch.tensor([3., 2., 1., 5.])
    weight1 = weight1.view(1, 1, 2, 2)
    net.conv1._parameters['weight'].data = weight1  # 自定义卷积核

    input = torch.tensor([[1., 2., 3.],  # 自定义输入
                          [4., 5., 6.],
                          [7., 8., 9.]])
    input = input.view(1, 1, 3, 3)
    output = net(input)
    print("前向传播输出：")
    print(output)
    print()

    # Loss Function
    target = torch.tensor(60.)
    criterion = nn.MSELoss()
    loss = criterion(output, target)
    print("MSE loss：", loss)
    print()

    # Backprop
    net.zero_grad()  # zeroes the gradient buffers of all parameters
    loss.backward()
    print("卷积核的梯度：")
    print(net.conv1.weight.grad)
    print()

    use_module = True
    if not use_module:
        # Update the weights     weight = weight - learning_rate * gradient
        learning_rate = 0.01
        for f in net.parameters():
            f.data.sub_(f.grad.data * learning_rate)
        print("手动更新")
        print(list(net.parameters()))
        """
        tensor([[[[2.5200, 1.3600],
                  [0.0400, 3.8800]]]], requires_grad=True)]
        """

    else:
        # However, as you use neural networks, you want to use various different update rules such as SGD,
        # Nesterov-SGD, Adam, RMSProp, etc. To enable this, we built a small package: torch.optim that
        # implements all these methods. Using it is very simple:
        import torch.optim as optim

        # create your optimizer
        optimizer = optim.SGD(net.parameters(), lr=0.01)

        # in your training loop:
        optimizer.zero_grad()  # zero the gradient buffers
        output = net(input)
        loss = criterion(output, target)
        loss.backward()
        optimizer.step()  # Does the update
        print("optim更新")
        print(list(net.parameters()))
        """
        tensor([[[[2.5200, 1.3600],
                  [0.0400, 3.8800]]]], requires_grad=True)]
        """

 返回目录

参考资料

《图解深度学习与神经网络：从张量到TensorFlow实现》_张平

https://pytorch.org/tutorials/beginner/blitz/neural_networks_tutorial.html#sphx-glr-beginner-blitz-neural-networks-tutorial-py

 返回目录