Pytorch 学习笔记（四）—— CNN卷积神经网络

卷积函数：

nn.functional.conv2d(input, weight, bias=None, stride=1, padding=0, dilation=1, groups=1)

对几个输入平面组成的输入信号应用一个2D卷积.
Parameters:
input – 形状为 (minibatch x in_channels x iH x iW) 的输入张量
weight – 形状为 (out_channels x in_channels/groups x kH x kW) 的滤波器
bias – 可选的偏置,形状为 (out_channels). 默认值: None
stride – 卷积核的步长. 可以是单个数字, 也可以是一个元组 (sH, sW). 默认值: 1
padding – 输入两端隐式零填充的个数. 可以是单个数字, 也可以是一个元组 (padH, padW). 默认值: 0
dilation – 卷积核中元素之间的空洞大小. 可以是单个数字, 也可以是一个元组 (dH, dW). 默认值: 1
groups – 将输入分成的组的个数. in_channels 的值要求能够被 groups 的值整除. 默认值: 1

池化函数：

nn.MaxPool2d(kernel_size, stride=None, padding=0, dilation=1, return_indices=False, ceil_mode=False)

对于多个输入通道组成的输入信号,应用二维的最大池化 max pooling 操作.
Parameters:
kernel_size – 最大池化操作时的窗口大小
stride – 最大池化操作时窗口移动的步长, 默认值是 kernel_size
padding – 输入的每条边隐式补0的数量
dilation – 用于控制窗口中元素的步长的参数
return_indices – 如果等于 True, 在返回 max pooling 结果的同时返回最大值的索引 这在之后的 Unpooling 时很有用
ceil_mode – 如果等于 True, 在计算输出大小时,将采用向上取整来代替默认的向下取整的方式

线性函数：

nn.functional.linear(input, weight, bias=None)

对输入的数据应用线性转换。
Shape:
Input: (N,∗,in_features)其中 * 表示任意数量的附加维度
Weight: (out_features,in_features)
Bias: (out_features)
Output: (N,∗,out_features)

import torch
import torch.nn as nn
import torch.utils.data as Data
import torchvision      # 数据库模块
import matplotlib.pyplot as plt

torch.manual_seed(1)    # reproducible

# Hyper Parameters
EPOCH = 1           # 训练整批数据多少次, 为了节约时间, 我们只训练一次
BATCH_SIZE = 50
LR = 0.001          # 学习率
DOWNLOAD_MNIST = True  # 如果你已经下载好了mnist数据就写上 False


# Mnist 手写数字
train_data = torchvision.datasets.MNIST(
    root='./mnist/',    # 保存或者提取位置
    train=True,  # this is training data
    transform=torchvision.transforms.ToTensor(),    # 转换 PIL.Image or numpy.ndarray 成
                                                    # torch.FloatTensor (C x H x W), 训练的时候 normalize 成 [0.0, 1.0] 区间
    download=DOWNLOAD_MNIST,          # 没下载就下载, 下载了就不用再下了
)

print(train_data.train_data.size())                 # (60000, 28, 28)
print(train_data.train_labels.size())               # (60000)
plt.imshow(train_data.train_data[0].numpy(), cmap='gray')
plt.title('%i' % train_data.train_labels[0])
plt.show()

训练数据的大小及标签大小

torch.Size([60000, 28, 28])
torch.Size([60000])

训练数据中的第一张图

设置训练数据，搭建CNN卷积神经网络

test_data = torchvision.datasets.MNIST(root='./mnist/', train=False)

# 批训练 50samples, 1 channel, 28x28 (50, 1, 28, 28)
train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True)

# 为了节约时间, 我们测试时只测试前2000个
test_x = torch.unsqueeze(test_data.test_data, dim=1).type(torch.FloatTensor)[:2000]/255.   # shape from (2000, 28, 28) to (2000, 1, 28, 28), value in range(0,1)
test_y = test_data.test_labels[:2000]

class CNN(nn.Module):
    def __init__(self):
        super(CNN, self).__init__()
        self.conv1 = nn.Sequential(  # input shape (1, 28, 28)
            nn.Conv2d(
                in_channels=1,      # input height
                out_channels=16,    # n_filters
                kernel_size=5,      # filter size
                stride=1,           # filter movement/step
                padding=2,          # 如果想要 con2d 出来的图片长宽没有变化, padding=(kernel_size-1)/2 当 stride=1
            ),                      # output shape (16, 28, 28)
            nn.ReLU(),              # activation
            nn.MaxPool2d(kernel_size=2),    # 在 2x2 空间里向下采样, output shape (16, 14, 14)
        )
        self.conv2 = nn.Sequential(  # input shape (16, 14, 14)
            nn.Conv2d(16, 32, 5, 1, 2),  # output shape (32, 14, 14)
            nn.ReLU(),  # activation
            nn.MaxPool2d(2),  # output shape (32, 7, 7)
        )
        self.out = nn.Linear(32 * 7 * 7, 10)   #全连接层 fully connected layer, output 10 classes

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = x.view(x.size(0), -1)   # 展平多维的卷积图成 (batch_size, 32 * 7 * 7)
        output = self.out(x)
        return output

cnn = CNN()
print(cnn)  # net architecture

输出网络结构：

CNN(
  (conv1): Sequential(
    (0): Conv2d(1, 16, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (conv2): Sequential(
    (0): Conv2d(16, 32, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (out): Linear(in_features=1568, out_features=10, bias=True)
)

训练过程

optimizer = torch.optim.Adam(cnn.parameters(), lr=LR)   # optimize all cnn parameters
loss_func = nn.CrossEntropyLoss()   # the target label is not one-hotted

# training and testing
for epoch in range(EPOCH):
    for step, (b_x, b_y) in enumerate(train_loader):   # 分配 batch data, normalize x when iterate train_loader
        output = cnn(b_x)               # cnn output
        loss = loss_func(output, b_y)   # cross entropy loss
        optimizer.zero_grad()           # clear gradients for this training step
        loss.backward()                 # backpropagation, compute gradients
        optimizer.step()                # apply gradients

test_output = cnn(test_x[:10])
pred_y = torch.max(test_output, 1)[1].data.numpy().squeeze()
print(pred_y, 'prediction number')
print(test_y[:10].numpy(), 'real number')

测试结果

[7 2 1 0 4 1 4 9 5 9] prediction number
[7 2 1 0 4 1 4 9 5 9] real number

2018.09.06 整理于莫烦Python教程
https://morvanzhou.github.io/tutorials/machine-learning/torch/4-01-CNN/