版权声明：本文为博主原创文章，欢迎转载，并请注明出处。联系方式：460356155@qq.com

VGGNet在2014年ImageNet图像分类任务竞赛中有出色的表现。网络结构如下图所示：

同样的，对32*32的CIFAR10图片，网络结构做了微调：删除了最后一层最大池化，具体参见网络定义代码，这里采用VGG19，并加入了BN：

 '''

 创建VGG块

 参数分别为输入通道数，输出通道数，卷积层个数，是否做最大池化

 '''

 def make_vgg_block(in_channel, out_channel, convs, pool=True):

     net = []

     # 不改变图片尺寸卷积

     net.append(nn.Conv2d(in_channel, out_channel, kernel_size=3, padding=1))

     net.append(nn.BatchNorm2d(out_channel))

     net.append(nn.ReLU(inplace=True))

     for i in range(convs - 1):

         # 不改变图片尺寸卷积

         net.append(nn.Conv2d(out_channel, out_channel, kernel_size=3, padding=1))

         net.append(nn.BatchNorm2d(out_channel))

         net.append(nn.ReLU(inplace=True))

     if pool:

         # 2*2最大池化，图片变为w/2 * h/2

         net.append(nn.MaxPool2d(2))

     return nn.Sequential(*net)

 # 定义网络模型

 class VGG19Net(nn.Module):

     def __init__(self):

         super(VGG19Net, self).__init__()

         net = []

         # 输入32*32，输出16*16

         net.append(make_vgg_block(3, 64, 2))

         # 输出8*8

         net.append(make_vgg_block(64, 128, 2))

         # 输出4*4

         net.append(make_vgg_block(128, 256, 4))

         # 输出2*2

         net.append(make_vgg_block(256, 512, 4))

         # 无池化层，输出保持2*2

         net.append(make_vgg_block(512, 512, 4, False))

         self.cnn = nn.Sequential(*net)

         self.fc = nn.Sequential(

             # 512个feature，每个feature 2*2

             nn.Linear(512*2*2, 256),

             nn.ReLU(),

             nn.Linear(256, 256),

             nn.ReLU(),

             nn.Linear(256, 10)

         )

     def forward(self, x):

         x = self.cnn(x)

         # x.size()[0]: batch size

         x = x.view(x.size()[0], -1)

         x = self.fc(x)

         return x

其余代码同深度学习识别CIFAR10：pytorch训练LeNet、AlexNet、VGG19实现及比较（一）运行结果如下：

Files already downloaded and verified
VGG19Net(
  (cnn): Sequential(
    (0): Sequential(
      (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace)
      (3): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (5): ReLU(inplace)
      (6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    )
    (1): Sequential(
      (0): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace)
      (3): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (5): ReLU(inplace)
      (6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    )
    (2): Sequential(
      (0): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace)
      (3): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (5): ReLU(inplace)
      (6): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (7): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (8): ReLU(inplace)
      (9): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (10): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (11): ReLU(inplace)
      (12): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    )
    (3): Sequential(
      (0): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace)
      (3): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (5): ReLU(inplace)
      (6): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (7): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (8): ReLU(inplace)
      (9): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (10): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (11): ReLU(inplace)
      (12): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    )
    (4): Sequential(
      (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace)
      (3): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (5): ReLU(inplace)
      (6): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (7): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (8): ReLU(inplace)
      (9): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
      (10): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (11): ReLU(inplace)
    )
  )
  (fc): Sequential(
    (0): Linear(in_features=2048, out_features=256, bias=True)
    (1): ReLU()
    (2): Linear(in_features=256, out_features=256, bias=True)
    (3): ReLU()
    (4): Linear(in_features=256, out_features=10, bias=True)
  )
)
Train Epoch: 1 [6400/50000 (13%)]    Loss: 1.991934  Acc: 22.000000
Train Epoch: 1 [12800/50000 (26%)]    Loss: 1.851721  Acc: 27.000000
Train Epoch: 1 [19200/50000 (38%)]    Loss: 1.765295  Acc: 31.000000
Train Epoch: 1 [25600/50000 (51%)]    Loss: 1.708027  Acc: 33.000000
Train Epoch: 1 [32000/50000 (64%)]    Loss: 1.652181  Acc: 36.000000
Train Epoch: 1 [38400/50000 (77%)]    Loss: 1.597727  Acc: 38.000000
Train Epoch: 1 [44800/50000 (90%)]    Loss: 1.552660  Acc: 41.000000
one epoch spend:  0:01:08.269581
EPOCH:1, ACC:55.08

Train Epoch: 2 [6400/50000 (13%)]    Loss: 1.139670  Acc: 60.000000
Train Epoch: 2 [12800/50000 (26%)]    Loss: 1.099960  Acc: 61.000000
Train Epoch: 2 [19200/50000 (38%)]    Loss: 1.078881  Acc: 62.000000
Train Epoch: 2 [25600/50000 (51%)]    Loss: 1.054403  Acc: 63.000000
Train Epoch: 2 [32000/50000 (64%)]    Loss: 1.031371  Acc: 64.000000
Train Epoch: 2 [38400/50000 (77%)]    Loss: 1.011668  Acc: 64.000000
Train Epoch: 2 [44800/50000 (90%)]    Loss: 0.995242  Acc: 65.000000
one epoch spend:  0:01:08.220392
EPOCH:2, ACC:71.01

Train Epoch: 3 [6400/50000 (13%)]    Loss: 0.823265  Acc: 71.000000
Train Epoch: 3 [12800/50000 (26%)]    Loss: 0.799878  Acc: 73.000000
Train Epoch: 3 [19200/50000 (38%)]    Loss: 0.791265  Acc: 73.000000
Train Epoch: 3 [25600/50000 (51%)]    Loss: 0.790027  Acc: 73.000000
Train Epoch: 3 [32000/50000 (64%)]    Loss: 0.777267  Acc: 73.000000
Train Epoch: 3 [38400/50000 (77%)]    Loss: 0.771953  Acc: 74.000000
Train Epoch: 3 [44800/50000 (90%)]    Loss: 0.766835  Acc: 74.000000
one epoch spend:  0:01:08.485721
EPOCH:3, ACC:69.48

Train Epoch: 4 [6400/50000 (13%)]    Loss: 0.640418  Acc: 78.000000
Train Epoch: 4 [12800/50000 (26%)]    Loss: 0.637256  Acc: 78.000000
Train Epoch: 4 [19200/50000 (38%)]    Loss: 0.631245  Acc: 79.000000
Train Epoch: 4 [25600/50000 (51%)]    Loss: 0.629215  Acc: 79.000000
Train Epoch: 4 [32000/50000 (64%)]    Loss: 0.625925  Acc: 79.000000
Train Epoch: 4 [38400/50000 (77%)]    Loss: 0.618307  Acc: 79.000000
Train Epoch: 4 [44800/50000 (90%)]    Loss: 0.617456  Acc: 79.000000
one epoch spend:  0:01:08.289673
EPOCH:4, ACC:77.2

Train Epoch: 5 [6400/50000 (13%)]    Loss: 0.537330  Acc: 82.000000
Train Epoch: 5 [12800/50000 (26%)]    Loss: 0.529751  Acc: 82.000000
Train Epoch: 5 [19200/50000 (38%)]    Loss: 0.529389  Acc: 82.000000
Train Epoch: 5 [25600/50000 (51%)]    Loss: 0.528106  Acc: 82.000000
Train Epoch: 5 [32000/50000 (64%)]    Loss: 0.526467  Acc: 82.000000
Train Epoch: 5 [38400/50000 (77%)]    Loss: 0.525133  Acc: 82.000000
Train Epoch: 5 [44800/50000 (90%)]    Loss: 0.521847  Acc: 82.000000
one epoch spend:  0:01:08.272084
EPOCH:5, ACC:78.26

Train Epoch: 6 [6400/50000 (13%)]    Loss: 0.435377  Acc: 85.000000
Train Epoch: 6 [12800/50000 (26%)]    Loss: 0.431456  Acc: 85.000000
Train Epoch: 6 [19200/50000 (38%)]    Loss: 0.443582  Acc: 85.000000
Train Epoch: 6 [25600/50000 (51%)]    Loss: 0.442819  Acc: 85.000000
Train Epoch: 6 [32000/50000 (64%)]    Loss: 0.443313  Acc: 85.000000
Train Epoch: 6 [38400/50000 (77%)]    Loss: 0.442025  Acc: 85.000000
Train Epoch: 6 [44800/50000 (90%)]    Loss: 0.441722  Acc: 85.000000
one epoch spend:  0:01:10.725170
EPOCH:6, ACC:80.91

Train Epoch: 7 [6400/50000 (13%)]    Loss: 0.350214  Acc: 88.000000
Train Epoch: 7 [12800/50000 (26%)]    Loss: 0.351490  Acc: 88.000000
Train Epoch: 7 [19200/50000 (38%)]    Loss: 0.361328  Acc: 88.000000
Train Epoch: 7 [25600/50000 (51%)]    Loss: 0.362231  Acc: 87.000000
Train Epoch: 7 [32000/50000 (64%)]    Loss: 0.364318  Acc: 87.000000
Train Epoch: 7 [38400/50000 (77%)]    Loss: 0.367137  Acc: 87.000000
Train Epoch: 7 [44800/50000 (90%)]    Loss: 0.375220  Acc: 87.000000
one epoch spend:  0:01:09.395538
EPOCH:7, ACC:80.55

Train Epoch: 8 [6400/50000 (13%)]    Loss: 0.297754  Acc: 90.000000
Train Epoch: 8 [12800/50000 (26%)]    Loss: 0.303383  Acc: 89.000000
Train Epoch: 8 [19200/50000 (38%)]    Loss: 0.305170  Acc: 89.000000
Train Epoch: 8 [25600/50000 (51%)]    Loss: 0.311823  Acc: 89.000000
Train Epoch: 8 [32000/50000 (64%)]    Loss: 0.309851  Acc: 89.000000
Train Epoch: 8 [38400/50000 (77%)]    Loss: 0.310422  Acc: 89.000000
Train Epoch: 8 [44800/50000 (90%)]    Loss: 0.312672  Acc: 89.000000
one epoch spend:  0:01:08.041167
EPOCH:8, ACC:80.54

Train Epoch: 9 [6400/50000 (13%)]    Loss: 0.277638  Acc: 90.000000
Train Epoch: 9 [12800/50000 (26%)]    Loss: 0.276622  Acc: 90.000000
Train Epoch: 9 [19200/50000 (38%)]    Loss: 0.276465  Acc: 90.000000
Train Epoch: 9 [25600/50000 (51%)]    Loss: 0.278001  Acc: 90.000000
Train Epoch: 9 [32000/50000 (64%)]    Loss: 0.277109  Acc: 90.000000
Train Epoch: 9 [38400/50000 (77%)]    Loss: 0.277029  Acc: 90.000000
Train Epoch: 9 [44800/50000 (90%)]    Loss: 0.275243  Acc: 90.000000
one epoch spend:  0:01:08.143754
EPOCH:9, ACC:83.53

Train Epoch: 10 [6400/50000 (13%)]    Loss: 0.205785  Acc: 92.000000
Train Epoch: 10 [12800/50000 (26%)]    Loss: 0.210659  Acc: 92.000000
Train Epoch: 10 [19200/50000 (38%)]    Loss: 0.214871  Acc: 92.000000
Train Epoch: 10 [25600/50000 (51%)]    Loss: 0.218910  Acc: 92.000000
Train Epoch: 10 [32000/50000 (64%)]    Loss: 0.220843  Acc: 92.000000
Train Epoch: 10 [38400/50000 (77%)]    Loss: 0.220417  Acc: 92.000000
Train Epoch: 10 [44800/50000 (90%)]    Loss: 0.221100  Acc: 92.000000
one epoch spend:  0:01:08.333929
EPOCH:10, ACC:79.01

Train Epoch: 11 [6400/50000 (13%)]    Loss: 0.186917  Acc: 93.000000
Train Epoch: 11 [12800/50000 (26%)]    Loss: 0.183512  Acc: 93.000000
Train Epoch: 11 [19200/50000 (38%)]    Loss: 0.182561  Acc: 93.000000
Train Epoch: 11 [25600/50000 (51%)]    Loss: 0.186446  Acc: 93.000000
Train Epoch: 11 [32000/50000 (64%)]    Loss: 0.187314  Acc: 93.000000
Train Epoch: 11 [38400/50000 (77%)]    Loss: 0.185967  Acc: 93.000000
Train Epoch: 11 [44800/50000 (90%)]    Loss: 0.189130  Acc: 93.000000
one epoch spend:  0:01:10.476138
EPOCH:11, ACC:81.57

Train Epoch: 12 [6400/50000 (13%)]    Loss: 0.136427  Acc: 95.000000
Train Epoch: 12 [12800/50000 (26%)]    Loss: 0.147904  Acc: 95.000000
Train Epoch: 12 [19200/50000 (38%)]    Loss: 0.154502  Acc: 94.000000
Train Epoch: 12 [25600/50000 (51%)]    Loss: 0.155767  Acc: 94.000000
Train Epoch: 12 [32000/50000 (64%)]    Loss: 0.158346  Acc: 94.000000
Train Epoch: 12 [38400/50000 (77%)]    Loss: 0.159562  Acc: 94.000000
Train Epoch: 12 [44800/50000 (90%)]    Loss: 0.159924  Acc: 94.000000
one epoch spend:  0:01:10.779635
EPOCH:12, ACC:84.38

Train Epoch: 13 [6400/50000 (13%)]    Loss: 0.110026  Acc: 96.000000
Train Epoch: 13 [12800/50000 (26%)]    Loss: 0.113738  Acc: 96.000000
Train Epoch: 13 [19200/50000 (38%)]    Loss: 0.117731  Acc: 96.000000
Train Epoch: 13 [25600/50000 (51%)]    Loss: 0.123653  Acc: 95.000000
Train Epoch: 13 [32000/50000 (64%)]    Loss: 0.127138  Acc: 95.000000
Train Epoch: 13 [38400/50000 (77%)]    Loss: 0.128938  Acc: 95.000000
Train Epoch: 13 [44800/50000 (90%)]    Loss: 0.131382  Acc: 95.000000
one epoch spend:  0:01:09.020651
EPOCH:13, ACC:83.46

Train Epoch: 14 [6400/50000 (13%)]    Loss: 0.122690  Acc: 96.000000
Train Epoch: 14 [12800/50000 (26%)]    Loss: 0.114584  Acc: 96.000000
Train Epoch: 14 [19200/50000 (38%)]    Loss: 0.122652  Acc: 96.000000
Train Epoch: 14 [25600/50000 (51%)]    Loss: 0.123031  Acc: 95.000000
Train Epoch: 14 [32000/50000 (64%)]    Loss: 0.123427  Acc: 95.000000
Train Epoch: 14 [38400/50000 (77%)]    Loss: 0.123146  Acc: 95.000000
Train Epoch: 14 [44800/50000 (90%)]    Loss: 0.124063  Acc: 95.000000
one epoch spend:  0:01:10.294790
EPOCH:14, ACC:82.27

Train Epoch: 15 [6400/50000 (13%)]    Loss: 0.087797  Acc: 97.000000
Train Epoch: 15 [12800/50000 (26%)]    Loss: 0.086152  Acc: 97.000000
Train Epoch: 15 [19200/50000 (38%)]    Loss: 0.088446  Acc: 97.000000
Train Epoch: 15 [25600/50000 (51%)]    Loss: 0.093510  Acc: 96.000000
Train Epoch: 15 [32000/50000 (64%)]    Loss: 0.092870  Acc: 96.000000
Train Epoch: 15 [38400/50000 (77%)]    Loss: 0.092416  Acc: 96.000000
Train Epoch: 15 [44800/50000 (90%)]    Loss: 0.095187  Acc: 96.000000
one epoch spend:  0:01:10.375479
EPOCH:15, ACC:82.73

Train Epoch: 16 [6400/50000 (13%)]    Loss: 0.066554  Acc: 97.000000
Train Epoch: 16 [12800/50000 (26%)]    Loss: 0.079139  Acc: 97.000000
Train Epoch: 16 [19200/50000 (38%)]    Loss: 0.078223  Acc: 97.000000
Train Epoch: 16 [25600/50000 (51%)]    Loss: 0.076825  Acc: 97.000000
Train Epoch: 16 [32000/50000 (64%)]    Loss: 0.079679  Acc: 97.000000
Train Epoch: 16 [38400/50000 (77%)]    Loss: 0.081081  Acc: 97.000000
Train Epoch: 16 [44800/50000 (90%)]    Loss: 0.081967  Acc: 97.000000
one epoch spend:  0:01:09.971818
EPOCH:16, ACC:85.45

Train Epoch: 17 [6400/50000 (13%)]    Loss: 0.061477  Acc: 98.000000
Train Epoch: 17 [12800/50000 (26%)]    Loss: 0.066804  Acc: 97.000000
Train Epoch: 17 [19200/50000 (38%)]    Loss: 0.069621  Acc: 97.000000
Train Epoch: 17 [25600/50000 (51%)]    Loss: 0.068841  Acc: 97.000000
Train Epoch: 17 [32000/50000 (64%)]    Loss: 0.069220  Acc: 97.000000
Train Epoch: 17 [38400/50000 (77%)]    Loss: 0.071493  Acc: 97.000000
Train Epoch: 17 [44800/50000 (90%)]    Loss: 0.070973  Acc: 97.000000
one epoch spend:  0:01:10.599626
EPOCH:17, ACC:83.02

Train Epoch: 18 [6400/50000 (13%)]    Loss: 0.095195  Acc: 96.000000
Train Epoch: 18 [12800/50000 (26%)]    Loss: 0.081690  Acc: 97.000000
Train Epoch: 18 [19200/50000 (38%)]    Loss: 0.076400  Acc: 97.000000
Train Epoch: 18 [25600/50000 (51%)]    Loss: 0.073249  Acc: 97.000000
Train Epoch: 18 [32000/50000 (64%)]    Loss: 0.072114  Acc: 97.000000
Train Epoch: 18 [38400/50000 (77%)]    Loss: 0.073739  Acc: 97.000000
Train Epoch: 18 [44800/50000 (90%)]    Loss: 0.073761  Acc: 97.000000
one epoch spend:  0:01:11.619880
EPOCH:18, ACC:83.67

Train Epoch: 19 [6400/50000 (13%)]    Loss: 0.049970  Acc: 98.000000
Train Epoch: 19 [12800/50000 (26%)]    Loss: 0.051812  Acc: 98.000000
Train Epoch: 19 [19200/50000 (38%)]    Loss: 0.053814  Acc: 98.000000
Train Epoch: 19 [25600/50000 (51%)]    Loss: 0.054168  Acc: 98.000000
Train Epoch: 19 [32000/50000 (64%)]    Loss: 0.054138  Acc: 98.000000
Train Epoch: 19 [38400/50000 (77%)]    Loss: 0.055356  Acc: 98.000000
Train Epoch: 19 [44800/50000 (90%)]    Loss: 0.055334  Acc: 98.000000
one epoch spend:  0:01:10.397104
EPOCH:19, ACC:84.23

Train Epoch: 20 [6400/50000 (13%)]    Loss: 0.059795  Acc: 98.000000
Train Epoch: 20 [12800/50000 (26%)]    Loss: 0.059780  Acc: 98.000000
Train Epoch: 20 [19200/50000 (38%)]    Loss: 0.060332  Acc: 98.000000
Train Epoch: 20 [25600/50000 (51%)]    Loss: 0.057949  Acc: 98.000000
Train Epoch: 20 [32000/50000 (64%)]    Loss: 0.056517  Acc: 98.000000
Train Epoch: 20 [38400/50000 (77%)]    Loss: 0.055322  Acc: 98.000000
Train Epoch: 20 [44800/50000 (90%)]    Loss: 0.053375  Acc: 98.000000
one epoch spend:  0:01:10.407573
EPOCH:20, ACC:84.51

CIFAR10 pytorch LeNet Train: EPOCH:20, BATCH_SZ:64, LR:0.01, ACC:85.45
train spend time:  0:23:45.010363

Process finished with exit code 0

准确率达到85%，对比AlexNet的75%，提升了10%。

深度学习识别CIFAR10：pytorch训练LeNet、AlexNet、VGG19实现及比较（三）的更多相关文章

  1. 深度学习识别CIFAR10：pytorch训练LeNet、AlexNet、VGG19实现及比较（二）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com AlexNet在2012年ImageNet图像分类任务竞赛中获得冠军.网络结构如下图所示: 对CIFA ...

  2. 深度学习识别CIFAR10：pytorch训练LeNet、AlexNet、VGG19实现及比较（一）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com 前面几篇文章介绍了MINIST,对这种简单图片的识别,LeNet-5可以达到99%的识别率. CIFA ...

  3. MINIST深度学习识别：python全连接神经网络和pytorch LeNet CNN网络训练实现及比较（三）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com 在前两篇文章MINIST深度学习识别:python全连接神经网络和pytorch LeNet CNN网 ...

  4. pytorch识别CIFAR10：训练ResNet-34（准确率80%）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com CNN的层数越多,能够提取到的特征越丰富,但是简单地增加卷积层数,训练时会导致梯度弥散或梯度爆炸. 何 ...

  5. 深度学习面试题12：LeNet(手写数字识别)

    目录 神经网络的卷积.池化.拉伸 LeNet网络结构 LeNet在MNIST数据集上应用 参考资料 LeNet是卷积神经网络的祖师爷LeCun在1998年提出,用于解决手写数字识别的视觉任务.自那时起 ...

  6. pytorch识别CIFAR10：训练ResNet-34（自定义transform，动态调整学习率，准确率提升到94.33%）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com 前面通过数据增强,ResNet-34残差网络识别CIFAR10,准确率达到了92.6. 这里对训练过程 ...

  7. pytorch识别CIFAR10：训练ResNet-34（数据增强，准确率提升到92.6%）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com 在前一篇中的ResNet-34残差网络,经过减小卷积核训练准确率提升到85%. 这里对训练数据集做数据 ...

  8. pytorch识别CIFAR10：训练ResNet-34（微调网络，准确率提升到85%）

    版权声明:本文为博主原创文章,欢迎转载,并请注明出处.联系方式:460356155@qq.com 在前一篇中的ResNet-34残差网络,经过训练准确率只达到80%. 这里对网络做点小修改,在最开始的 ...

  9. 用CNTK搞深度学习 （二） 训练基于RNN的自然语言模型 ( language model )

    前一篇文章  用 CNTK 搞深度学习 (一) 入门    介绍了用CNTK构建简单前向神经网络的例子.现在假设读者已经懂得了使用CNTK的基本方法.现在我们做一个稍微复杂一点,也是自然语言挖掘中很火 ...

随机推荐

  1. Linux知识要点大全（第一章）

    第一章   Linux操作系统简介 主要内容: 1:Linux的发展历史               2:开源软件               3:Linux系统结构与特性               ...

  2. SQL Server的case when用法

    1.简单sql一例 SELECT top 10 CASE WHEN IDENTITY_ID='1' THEN '管理员' WHEN IDENTITY_ID='5' THEN '学生' ELSE '无' ...

  3. c# 获取当前时间的微秒

    获取毫秒大家都经常用到. 大家应该都知道怎么用. 但是,毫秒下面还有微秒. 其实这个方法也已经在c#中. 只不过很少有人用到,所以查找资料也很少有人说. 下面代码就是获取微秒的方式: DateTime ...

  4. 深度学习之卷积神经网络(CNN)详解与代码实现（二）

    用Tensorflow实现卷积神经网络(CNN) 本文系作者原创,转载请注明出处:https://www.cnblogs.com/further-further-further/p/10737065. ...

  5. 企业级自动化运维工具---puppet详解

    本文收录在Linux运维企业架构实战系列 1.认识puppet 1.1 引入 puppet是什么,咱们先不用专业的名词解释它,咱们先描述一些工作场景,看明白这些工作场景,自然会知道puppet是什么. ...

  6. cocos creator主程入门教程（十）—— A*寻路

    摘要: 五邑隐侠,本名关健昌,10年游戏生涯,现隐居五邑.本系列文章以TypeScript为介绍语言. 这一篇介绍A*寻路算法.在RPG.SLG.模拟经营类游戏,有需要给角色寻路的需求,一般寻路我们采 ...

  7. DevOps概述

    Devops概念 转载自 devops实践-开篇感想 DevOps(英文Development和Operations的组合)是一组过程.方法与系统的统称,用于促进开发(应用程序/软件工程).技术运营和 ...

  8. 代理模式 PROXY Surrogate 结构型 设计模式（十四）

    代理模式 PROXY 别名Surrogate 意图 为其他的对象提供一种代理以控制对这个对象的访问. 代理模式含义比较清晰,就是中间人,中介公司,经纪人... 在计算机程序中,代理就表示一个客户端不想 ...

  9. 15个常用的javaScript正则表达式

    1 用户名正则 //用户名正则,4到16位(字母,数字,下划线,减号) ,}$/; //输出 true console.log(uPattern.test("iFat3")); 2 ...

  10. springMVC实现增删改查

    首先需要准备好一张数据库表我这里用emp这张表:具体代码: /* SQLyog 企业版 - MySQL GUI v8.14 MySQL - 5.1.73-community ************* ...