代码：
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 = False # 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 ) #plot one example # print(train_data.test_data.shape)#torch.Size([60000, 28, 28]) # print(train_data.train_labels.shape)#torch.Size([60000]) # print(train_data.train_data[0].shape)#torch.Size([28, 28]) # # plt.imshow(train_data.train_data[1],cmap='gray') # plt.title('%d'%train_data.train_labels[1]) # plt.show() #测试数据 test_data = torchvision.datasets.MNIST(root='./mnist/', train=False) # print(test_data.test_data.shape)#torch.Size([10000, 28, 28]) # 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] # 批训练 50samples, 1 channel, 28x28 (50, 1, 28, 28) train_loader = Data.DataLoader(dataset=train_data, batch_size=BATCH_SIZE, shuffle=True) class CNN(nn.Module): def __init__(self): super(CNN, self).__init__() self.conv1=nn.Sequential( nn.Conv2d( in_channels=1, 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(), nn.MaxPool2d(kernel_size=2)# output shape (16, 14, 14) ) self.conv2=nn.Sequential( nn.Conv2d(16,32,5,1,2),# output shape (32, 14, 14) nn.ReLU(), nn.MaxPool2d(2)# output shape (32, 7, 7) ) self.out=nn.Linear(32*7*7,10)# fully connected layer, output 10 classes def forward(self, x): x=self.conv1(x) x=self.conv2(x) #print(x.shape)#output:torch.Size([50, 32, 7, 7]) x = x.view(x.size(0), -1) # 展平多维的卷积图成 (batch_size, 32 * 7 * 7) # print(x.shape)#output:torch.Size([50, 1568]) output = self.out(x) return output cnn=CNN() 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 print('step:',step) 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]) #test_x[:10].shape=torch.Size([10, 1, 28, 28]) #test_output.shape=torch.Size([10, 10]) print('test_output：',test_output) # test_output： tensor([[-1383.2828, -1148.1272, 311.1780, 153.0877, -3062.3340, -886.6730, # -5819.7256, 3619.9558, -1544.4225, 193.6745], # [ 282.6339, 647.2642, 3027.1570, -379.0817, -3403.5310, -2406.4951, # -1117.4684, -4085.4429, -306.6578, -3844.1602], # [-1329.7642, 1895.3890, -755.7719, -1378.9316, -314.2351, -1607.4249, # -1026.8795, -428.1658, -385.1328, -1404.5205], # [ 2991.5627, -3583.5374, -554.1349, -2472.6204, -1712.7700, -1092.7367, # 148.9156, -1580.6696, -1126.8331, -477.7481], # [-1818.9655, -1502.3574, -1620.6603, -2142.3472, 2529.0496, -2008.2731, # -1585.5699, -786.7817, -1372.2627, 848.0875], # [-1415.7609, 2248.9607, -909.5534, -1656.6108, -311.2874, -2255.2163, # -1643.2495, -149.4040, -342.9626, -1372.8961], # [-3766.0422, -484.8116, -1971.9016, -2483.8538, 1448.3118, -1048.7388, # -2411.9790, -1089.5471, 422.1722, 249.8736], # [-2933.3752, -877.4833, -671.7119, -573.4670, 63.9295, -497.9561, # -2236.4597, -1218.2463, -296.5850, 1256.0739], # [-2187.7292, -4899.0063, -2404.6597, -2595.0764, -2987.9624, 2052.1494, # 335.9461, -2942.6995, 275.7964, -551.2797], # [-1903.9233, -3449.5530, -1652.7020, -1087.9016, -515.1445, -1170.5551, # -3734.2666, 628.9314, 69.0235, 2096.6257]], # grad_fn=<AddmmBackward>) print('test_output.shape：',test_output.shape) # test_output.shape： torch.Size([10, 10]) pred_y = torch.max(test_output, 1)[1].data.numpy().squeeze() print(pred_y, 'prediction number') print(test_y[:10].numpy(), 'real number')