使用pytorch搭建简单CNN

下载安装

pytorch官网上给出了通过pip，conda安装的命令，由于国内网络的原因，pip需要镜像。有CUDA版本和CPU版本的。安装anaconda可以方便地使用许多独立的python环境，在pycharm里可以直接切换使用的conda环境。直接使用conda安装pytorch不需要翻墙，会下载torch和torchvison，torchvision主要是跟图像有关的库，有训练的数据集和训练好的模型等。

使用CNN训练

定义几个全局变量，epoch是训练的次数，因为我用的cpu训练很慢就设为1了，batch_size是分批训练，每批的数据个数，lr是学习率，这里使用的是MNIST数据，如果之前没有下载过，就把这个设成True，如果已经下载过就设成False。

EPOCH = 1            
BATCH_SIZE = 50
LR = 0.001
DOWNLOAD_MNIST = False

train是使用数据集的training data，然后将其转换为tensor（CWH）的形式，如果下载过就不用下载了。

train_data = torchvision.datasets.MNIST(
    root='./mnist/',
    train=True,                                     
    transform=torchvision.transforms.ToTensor(),
    download=DOWNLOAD_MNIST,
)

有60000个训练数据，是28*28像素的灰度图。

把数据放到dataloader里，以便进行批训练，shuffle参数是是否打乱顺序。

train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True)

搭建神经网络，首先是一个卷积层，接一个ReLU激活函数，接一个池化层，卷积层将维度提升到了16维，padding是(kernel_size-1)/2这样保持图片大小不变，因此一开始是12828卷积之后是162828，池化之后是161414。之后再连接的卷积层有32个filter，padding还是2，因此变为321414，池化之后变为3277。最后输出层要把它拉直作为一个线性的变量去求。

class CNN(nn.Module):
    def __init__(self):
        super(CNN, self).__init__()
        self.conv1 = nn.Sequential(         
            nn.Conv2d(
                in_channels=1,              
                out_channels=16,            
                kernel_size=5,              
                stride=1,                   
                padding=2,                  
            ),                              
            nn.ReLU(),                      
            nn.MaxPool2d(kernel_size=2),    
        )
        self.conv2 = nn.Sequential(         
            nn.Conv2d(16, 32, 5, 1, 2),     
            nn.ReLU(),                      
            nn.MaxPool2d(2),                
        )
        self.out = nn.Linear(32 * 7 * 7, 10)  

使用Adam算法加速，损失函数用交叉熵来求。

optimizer = torch.optim.Adam(cnn.parameters(), lr=LR)   
loss_func = nn.CrossEntropyLoss()    

求损失函数，然后优化器将这一步的梯度清零，反向传播，然后按照梯度进行移动。

for epoch in range(EPOCH):
    for step, (b_x, b_y) in enumerate(train_loader):   
        output = cnn(b_x)[0]               
        loss = loss_func(output, b_y)   
        optimizer.zero_grad()           
        loss.backward()                 
        optimizer.step()  

完整代码：

import os
import torch
import torch.nn as nn
import torch.utils.data as Data
import torchvision
import matplotlib.pyplot as plt

# torch.manual_seed(1)

EPOCH = 1
BATCH_SIZE = 50
LR = 0.001
DOWNLOAD_MNIST = False


if not(os.path.exists('./mnist/')) or not os.listdir('./mnist/'):
    DOWNLOAD_MNIST = True

train_data = torchvision.datasets.MNIST(
    root='./mnist/',
    train=True,
    transform=torchvision.transforms.ToTensor(),
    download=DOWNLOAD_MNIST,
)

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()

train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True)

test_data = torchvision.datasets.MNIST(root='./mnist/', train=False)
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(
            nn.Conv2d(
                in_channels=1,
                out_channels=16,
                kernel_size=5,
                stride=1,
                padding=2,
            ),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=2),
        )
        self.conv2 = nn.Sequential(
            nn.Conv2d(16, 32, 5, 1, 2),
            nn.ReLU(),
            nn.MaxPool2d(2),
        )
        self.out = nn.Linear(32 * 7 * 7, 10)

    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = x.view(x.size(0), -1)           # flatten the output of conv2 to (batch_size, 32 * 7 * 7)
        output = self.out(x)
        return output, x


cnn = CNN()
print(cnn)

optimizer = torch.optim.Adam(cnn.parameters(), lr=LR)
loss_func = nn.CrossEntropyLoss()


from matplotlib import cm
try: from sklearn.manifold import TSNE; HAS_SK = True
except: HAS_SK = False; print('Please install sklearn for layer visualization')
def plot_with_labels(lowDWeights, labels):
    plt.cla()
    X, Y = lowDWeights[:, 0], lowDWeights[:, 1]
    for x, y, s in zip(X, Y, labels):
        c = cm.rainbow(int(255 * s / 9)); plt.text(x, y, s, backgroundcolor=c, fontsize=9)
    plt.xlim(X.min(), X.max()); plt.ylim(Y.min(), Y.max()); plt.title('Visualize last layer'); plt.show(); plt.pause(0.01)

plt.ion()

for epoch in range(EPOCH):
    for step, (b_x, b_y) in enumerate(train_loader):
        output = cnn(b_x)[0]
        loss = loss_func(output, b_y)
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        if step % 50 == 0:
            test_output, last_layer = cnn(test_x)
            pred_y = torch.max(test_output, 1)[1].data.numpy()
            accuracy = float((pred_y == test_y.data.numpy()).astype(int).sum()) / float(test_y.size(0))
            print('Epoch: ', epoch, '| train loss: %.4f' % loss.data.numpy(), '| test accuracy: %.2f' % accuracy)
            if HAS_SK:
                # Visualization of trained flatten layer (T-SNE)
                tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
                plot_only = 500
                low_dim_embs = tsne.fit_transform(last_layer.data.numpy()[:plot_only, :])
                labels = test_y.numpy()[:plot_only]
                plot_with_labels(low_dim_embs, labels)
plt.ioff()

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

训练的正确率最终能提高到98。