P24_现有网络模型的使用及修改

24.1 VGG16网络模型：

（1）打开pytorch(0.9.0)—torchvision.models—VGG

（2）参数

点击查看代码

pretrained (bool) – If True, returns a model pre-trained on ImageNet
progress (bool) – If True, displays a progress bar of the download to stderr

24.2 ImageNet数据集

（1）ImageNet数据集要求：

This requires scipy to be installed

【可以在pycharm的终端输入：pip list，查看是否安装了scipy库】

（2）下载该数据集

点击查看代码

import torchvision.datasets
train_data = torchvision.datasets.ImageNet("./data_image_net",split='train',                                      download=True,transform=torchvision.transforms.ToTensor())

（3）报错如下：

点击查看代码

The archive ILSVRC2012_devkit_t12.tar.gz is not present in the root directory or is corrupted. You need to download it externally and place it in ./data_image_net.

【正在尝试使用的 torchvision.datasets.ImageNet 数据集没有自动下载成功】

（4）PyTorch 要求用户必须手动下载 ImageNet 数据集

ImageNet数据集太大了，仅训练集就有147.9GB

所以：就没有办法通过下载ImageNet数据集去验证vgg16模型的pretrained这个参数：
pretrained (bool) – If True, returns a model pre-trained on ImageNet

24.3 改变现有网络的参数：添加or修改

（1）查看vgg16不同pretrained情况中其网络模型参数的差别

①把pretrained这个参数设置为true或false的情况下，看其下载的网络模型其中的参数有什么差别

点击查看代码

vgg16_false = torchvision.models.vgg16(pretrained=False)
vgg16_true = torchvision.models.vgg16(pretrained=True)
print("ok")

②把断点打在print("ok")，让其debug到print的上一行代码

vgg16_true和vgg16_false:

打开vgg16_true的classifier的Protected Attributes中的_modules下的'0'中的权重weight

打开vgg16_false的classifier的Protected Attributes中的_modules下的'0'中的权重weight

③【区别：vgg16_false没训练，vgg16_true训练好的】

（2）print(vgg16_true)，查看vgg16的网络架构

点击查看代码

VGG((features): Sequential((0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(1): ReLU(inplace=True)(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(3): ReLU(inplace=True)(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(6): ReLU(inplace=True)(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(8): ReLU(inplace=True)(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(11): ReLU(inplace=True)(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(13): ReLU(inplace=True)(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(15): ReLU(inplace=True)(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(18): ReLU(inplace=True)(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(20): ReLU(inplace=True)(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(22): ReLU(inplace=True)(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(25): ReLU(inplace=True)(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(27): ReLU(inplace=True)(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(29): ReLU(inplace=True)(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False))(avgpool): AdaptiveAvgPool2d(output_size=(7, 7))(classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=1000, bias=True))
)

通过out_features=1000，可以知道vgg16能分出的类别是1000个

（3）应用vgg16模型到CIFAR10数据集

vgg16模型能分出的类别是1000个，CIFAR10数据集把数据分为10类

①法一：将vgg16模型架构中的最后一行代码中的：(6): Linear(in_features=4096, out_features=1000, bias=True)里面的out_features=1000改成out_features=10

②法二：在线性层 (6): Linear(in_features=4096, out_features=1000, bias=True)后面再加一层：使in_features为1000，out_features为10

（4）【添加】使用（3）的法二：add_module

①输入：

点击查看代码

import torchvision.datasets# train_data = torchvision.datasets.ImageNet("./data_image_net",split='train',
#                                            download=True,transform=torchvision.transforms.ToTensor())
from torch import nnvgg16_false = torchvision.models.vgg16(pretrained=False)
vgg16_true = torchvision.models.vgg16(pretrained=True)
print(vgg16_true)train_data = torchvision.datasets.CIFAR10("./dataset",train=True,transform=torchvision.transforms.ToTensor(),download=True)
vgg16_true.add_module("add_linear",nn.Linear(in_features=1000,out_features=10))
print(vgg16_true)

②得到的结果中就会多加一层add_linear层：

(add_linear): Linear(in_features=1000, out_features=10, bias=True)

点击查看代码

Files already downloaded and verified
VGG((features): Sequential((0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(1): ReLU(inplace=True)(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(3): ReLU(inplace=True)(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(6): ReLU(inplace=True)(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(8): ReLU(inplace=True)(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(11): ReLU(inplace=True)(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(13): ReLU(inplace=True)(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(15): ReLU(inplace=True)(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(18): ReLU(inplace=True)(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(20): ReLU(inplace=True)(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(22): ReLU(inplace=True)(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(25): ReLU(inplace=True)(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(27): ReLU(inplace=True)(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(29): ReLU(inplace=True)(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False))(avgpool): AdaptiveAvgPool2d(output_size=(7, 7))(classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=1000, bias=True))(add_linear): Linear(in_features=1000, out_features=10, bias=True)
)

③想要(add_linear): Linear(in_features=1000, out_features=10, bias=True)加到classifier()里面，而不是整个VGG()里面，可以修改代码：

vgg16_true.classifier.add_module("add_linear",nn.Linear(in_features=1000,out_features=10))

可见：add_linear已经加到了classifier里面：

点击查看代码

  (classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=1000, bias=True)(add_linear): Linear(in_features=1000, out_features=10, bias=True))

（5）【修改】使用（3）的法一：[i]

由于vgg16_true已经被修改过（即被添加了add_linear层），现在在vgg16_false进行修改

①先打印vgg16_false

print(vgg16_false)

得到classifier里面：

点击查看代码

  (classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=1000, bias=True))

②直接修改：

vgg16_false.classifier[6] = nn.Linear(in_features=4096,out_features=10,bias=True)

得到：

点击查看代码

  (classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=10, bias=True))

可见，out_features变成了10