第P6周：VGG-16算法-Pytorch实现人脸识别
- 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
- 🍖 原作者：K同学啊
-
要求：
- 保存训练过程中的最佳模型权重
- 调用官方的VGG-16网络框架
-
拔高（可选）：
- 测试集准确率达到60%（难度有点大，但是这个过程可以学到不少）
- 手动搭建VGG-16
-
我的环境：
- 操作系统：CentOS7
- 显卡：RTX3090
- 显卡驱动：535.154.05
- CUDA版本: 12.2
- 语言环境：Python3.10
- 编译器：Jupyter Lab
- 深度学习环境：
- torch==12.1
- torchvision==0.18.1
一、前期准备
1. GPU 设置
import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os,PIL,pathlib,warnings
warnings.filterwarnings("ignore") #忽略警告信息
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
#device = "cpu"
device
device(type='cuda')
2. 导入数据
import os,PIL,random,pathlib
data_dir = './data/'
#data_dir = './small_data/'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("/")[1] for path in data_paths]
classeNames
['Hugh Jackman', 'Jennifer Lawrence', 'Nicole Kidman', 'Johnny Depp', 'Megan Fox', 'Will Smith', 'Angelina Jolie', 'Tom Hanks', 'Tom Cruise', 'Sandra Bullock', 'Denzel Washington', 'Kate Winslet', 'Natalie Portman', 'Scarlett Johansson', 'Brad Pitt', 'Robert Downey Jr', 'Leonardo DiCaprio']
# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
train_transforms = transforms.Compose([
transforms.Resize([224, 224]), # 将输入图片resize成统一尺寸
transforms.RandomHorizontalFlip(), # 随机水平翻转
transforms.ToTensor(), # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
transforms.Normalize( # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]) # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])
#total_data = datasets.ImageFolder("./small_data/",transform=train_transforms)
total_data = datasets.ImageFolder("./data/",transform=train_transforms)
total_data
Dataset ImageFolder
Number of datapoints: 1800
Root location: ./data/
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=True)
RandomHorizontalFlip(p=0.5)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
total_data.class_to_idx
{'Angelina Jolie': 0,
'Brad Pitt': 1,
'Denzel Washington': 2,
'Hugh Jackman': 3,
'Jennifer Lawrence': 4,
'Johnny Depp': 5,
'Kate Winslet': 6,
'Leonardo DiCaprio': 7,
'Megan Fox': 8,
'Natalie Portman': 9,
'Nicole Kidman': 10,
'Robert Downey Jr': 11,
'Sandra Bullock': 12,
'Scarlett Johansson': 13,
'Tom Cruise': 14,
'Tom Hanks': 15,
'Will Smith': 16}
3. 数据集划分
train_size = int(0.8 * len(total_data))
test_size = len(total_data) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(total_data, [train_size, test_size])
train_dataset, test_dataset
(<torch.utils.data.dataset.Subset at 0x7fdee45473a0>,
<torch.utils.data.dataset.Subset at 0x7fdee45470a0>)
#batch_size = 32
batch_size = 16
train_dl = torch.utils.data.DataLoader(train_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=1)
test_dl = torch.utils.data.DataLoader(test_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=1)
for X, y in test_dl:
print("Shape of X [N, C, H, W]: ", X.shape)
print("Shape of y: ", y.shape, y.dtype)
break
Shape of X [N, C, H, W]: torch.Size([16, 3, 224, 224])
Shape of y: torch.Size([16]) torch.int64
二、调用官方VGG-16模型
VGG-16（Visual Geometry Group-16）是由牛津大学视觉几何组（Visual Geometry Group）提出的一种深度卷积神经网络架构，用于图像分类和对象识别任务。VGG-16在2014年被提出，是VGG系列中的一种。VGG-16之所以备受关注，是因为它在ImageNet图像识别竞赛中取得了很好的成绩，展示了其在大规模图像识别任务中的有效性。
以下是VGG-16的主要特点：
- 深度：VGG-16由16个卷积层和3个全连接层组成，因此具有相对较深的网络结构。这种深度有助于网络学习到更加抽象和复杂的特征。
- 卷积层的设计：VGG-16的卷积层全部采用3x3的卷积核和步长为1的卷积操作，同时在卷积层之后都接有ReLU激活函数。这种设计的好处在于，通过堆叠多个较小的卷积核，可以提高网络的非线性建模能力，同时减少了参数数量，从而降低了过拟合的风险。
- 池化层：在卷积层之后，VGG-16使用最大池化层来减少特征图的空间尺寸，帮助提取更加显著的特征并减少计算量。
- 全连接层：VGG-16在卷积层之后接有3个全连接层，最后一个全连接层输出与类别数相对应的向量，用于进行分类。
VGG-16结构说明：
- 13个卷积层（Convolutional Layer），分别用blockX_convX表示；
- 3个全连接层（Fully connected Layer），用classifier表示；
- 5个池化层（Pool layer）。
VGG-16包含了16个隐藏层（13个卷积层和3个全连接层），故称为VGG-16
from torchvision.models import vgg16
device = "cuda" if torch.cuda.is_available() else "cpu"
#device = "cpu"
print("Using {} device".format(device))
# 加载预训练模型，并且对模型进行微调
model = vgg16(pretrained = True).to(device) # 加载预训练的vgg16模型
for param in model.parameters():
param.requires_grad = False # 冻结模型的参数，这样子在训练的时候只训练最后一层的参数
# 修改classifier模块的第6层（即：(6): Linear(in_features=4096, out_features=2, bias=True)）
# 注意查看我们下方打印出来的模型
model.classifier._modules['6'] = nn.Linear(4096,len(classeNames)) # 修改vgg16模型中最后一层全连接层，输出目标类别个数
# 修改模型全连接层参数
model.classifier._modules['0'] = nn.Linear(25088,1024)
model.classifier._modules['3'] = nn.Linear(1024,128)
model.classifier._modules['6'] = nn.Linear(128,len(classeNames)) # 修改vgg16模型中最后一层全连接层，输出目标类别个数
model.to(device)
model
Using cuda device
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=1024, bias=True)
(1): ReLU(inplace=True)
(2): Dropout(p=0.5, inplace=False)
(3): Linear(in_features=1024, out_features=128, bias=True)
(4): ReLU(inplace=True)
(5): Dropout(p=0.5, inplace=False)
(6): Linear(in_features=128, out_features=17, bias=True)
)
)
import torchsummary as summary
summary.summary(model, input_size=(3, 224, 224))
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [-1, 64, 224, 224] 1,792
ReLU-2 [-1, 64, 224, 224] 0
Conv2d-3 [-1, 64, 224, 224] 36,928
ReLU-4 [-1, 64, 224, 224] 0
MaxPool2d-5 [-1, 64, 112, 112] 0
Conv2d-6 [-1, 128, 112, 112] 73,856
ReLU-7 [-1, 128, 112, 112] 0
Conv2d-8 [-1, 128, 112, 112] 147,584
ReLU-9 [-1, 128, 112, 112] 0
MaxPool2d-10 [-1, 128, 56, 56] 0
Conv2d-11 [-1, 256, 56, 56] 295,168
ReLU-12 [-1, 256, 56, 56] 0
Conv2d-13 [-1, 256, 56, 56] 590,080
ReLU-14 [-1, 256, 56, 56] 0
Conv2d-15 [-1, 256, 56, 56] 590,080
ReLU-16 [-1, 256, 56, 56] 0
MaxPool2d-17 [-1, 256, 28, 28] 0
Conv2d-18 [-1, 512, 28, 28] 1,180,160
ReLU-19 [-1, 512, 28, 28] 0
Conv2d-20 [-1, 512, 28, 28] 2,359,808
ReLU-21 [-1, 512, 28, 28] 0
Conv2d-22 [-1, 512, 28, 28] 2,359,808
ReLU-23 [-1, 512, 28, 28] 0
MaxPool2d-24 [-1, 512, 14, 14] 0
Conv2d-25 [-1, 512, 14, 14] 2,359,808
ReLU-26 [-1, 512, 14, 14] 0
Conv2d-27 [-1, 512, 14, 14] 2,359,808
ReLU-28 [-1, 512, 14, 14] 0
Conv2d-29 [-1, 512, 14, 14] 2,359,808
ReLU-30 [-1, 512, 14, 14] 0
MaxPool2d-31 [-1, 512, 7, 7] 0
AdaptiveAvgPool2d-32 [-1, 512, 7, 7] 0
Linear-33 [-1, 1024] 25,691,136
ReLU-34 [-1, 1024] 0
Dropout-35 [-1, 1024] 0
Linear-36 [-1, 128] 131,200
ReLU-37 [-1, 128] 0
Dropout-38 [-1, 128] 0
Linear-39 [-1, 17] 2,193
================================================================
Total params: 40,539,217
Trainable params: 25,824,529
Non-trainable params: 14,714,688
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 218.61
Params size (MB): 154.64
Estimated Total Size (MB): 373.83
----------------------------------------------------------------
- 使用PyTorch编写VGG-16模型
import torch.nn.functional as F
class VGG16Model(nn.Module):
def __init__(self, nclasses=10):
super(VGG16Model, self).__init__()
self.conv11=nn.Sequential(
nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1), ## 224*224*64
nn.BatchNorm2d(64),
nn.ReLU())
self.conv12=nn.Sequential(
nn.Conv2d(64, 64, kernel_size=3, stride=1, padding=1), ## 224*224*64
nn.BatchNorm2d(64),
nn.ReLU())
self.pool1=nn.Sequential(
nn.MaxPool2d(2)) ## 112*112*64
self.conv21=nn.Sequential(
nn.Conv2d(64, 128, kernel_size=3, padding=1), ## 112*112*128
nn.BatchNorm2d(128),
nn.ReLU())
self.conv22=nn.Sequential(
nn.Conv2d(128, 128, kernel_size=3, padding=1), ## 112*112*128
nn.BatchNorm2d(128),
nn.ReLU())
self.pool2=nn.Sequential(
nn.MaxPool2d(2)) ## 56*56*128
self.conv31=nn.Sequential(
nn.Conv2d(128, 256, kernel_size=3, padding=1), ## 56*56*256
nn.BatchNorm2d(256),
nn.ReLU())
self.conv32=nn.Sequential(
nn.Conv2d(256, 256, kernel_size=3, padding=1), ## 56*56*256
nn.BatchNorm2d(256),
nn.ReLU())
self.conv33=nn.Sequential(
nn.Conv2d(256, 256, kernel_size=3, padding=1), ## 56*56*256
nn.BatchNorm2d(256),
nn.ReLU())
self.pool3=nn.Sequential(
nn.MaxPool2d(2)) ## 28*28*256
self.conv41=nn.Sequential(
nn.Conv2d(256, 512, kernel_size=3, padding=1), ## 28*28*512
nn.BatchNorm2d(512),
nn.ReLU())
self.conv42=nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, padding=1), ## 28*28*512
nn.BatchNorm2d(512),
nn.ReLU())
self.conv43=nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, padding=1), ## 28*28*512
nn.BatchNorm2d(512),
nn.ReLU())
self.pool4=nn.Sequential(
nn.MaxPool2d(2)) ## 14*14*512
self.conv51=nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, padding=1), ## 14*14*512
nn.BatchNorm2d(512),
nn.ReLU())
self.conv52=nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, padding=1), ## 14*14*512
nn.BatchNorm2d(512),
nn.ReLU())
self.conv53=nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, padding=1), ## 14*14*512
nn.BatchNorm2d(512),
nn.ReLU())
self.pool5=nn.Sequential(
nn.MaxPool2d(2)) ## 7*7*512
self.fc1 = nn.Sequential(
nn.Dropout(0.5),
nn.Linear(7*7*512, 4096),
nn.ReLU())
self.fc2 = nn.Sequential(
nn.Dropout(0.5),
nn.Linear(4096, 4096),
nn.ReLU())
self.fc3 = nn.Sequential(
nn.Linear(4096, nclasses))
def forward(self, x):
batch_size = x.size(0)
x = self.conv11(x)
x = self.conv12(x)
x = self.pool1(x)
x = self.conv21(x)
x = self.conv22(x)
x = self.pool2(x)
x = self.conv31(x)
x = self.conv32(x)
x = self.conv33(x)
x = self.pool3(x)
x = self.conv41(x)
x = self.conv42(x)
x = self.conv43(x)
x = self.pool4(x)
x = self.conv51(x)
x = self.conv52(x)
x = self.conv53(x)
x = self.pool5(x)
x = x.view(batch_size, -1)
x = self.fc1(x)
x = self.fc2(x)
x = self.fc3(x)
return x
device = "cuda" if torch.cuda.is_available() else "cpu"
#device = "cpu"
print("Using {} device".format(device))
vgg16_model = VGG16Model(len(classeNames)).to(device)
vgg16_model
Using cuda device
VGG16Model(
(conv11): 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()
)
(conv12): Sequential(
(0): Conv2d(64, 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()
)
(pool1): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(conv21): 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()
)
(conv22): Sequential(
(0): Conv2d(128, 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()
)
(pool2): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(conv31): 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()
)
(conv32): Sequential(
(0): Conv2d(256, 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()
)
(conv33): Sequential(
(0): Conv2d(256, 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()
)
(pool3): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(conv41): 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()
)
(conv42): 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()
)
(conv43): 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()
)
(pool4): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(conv51): 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()
)
(conv52): 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()
)
(conv53): 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()
)
(pool5): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(fc1): Sequential(
(0): Dropout(p=0.5, inplace=False)
(1): Linear(in_features=25088, out_features=4096, bias=True)
(2): ReLU()
)
(fc2): Sequential(
(0): Dropout(p=0.5, inplace=False)
(1): Linear(in_features=4096, out_features=4096, bias=True)
(2): ReLU()
)
(fc3): Sequential(
(0): Linear(in_features=4096, out_features=17, bias=True)
)
)
import torchsummary as summary
summary.summary(vgg16_model, (3, 224, 224))
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [-1, 64, 224, 224] 1,792
BatchNorm2d-2 [-1, 64, 224, 224] 128
ReLU-3 [-1, 64, 224, 224] 0
Conv2d-4 [-1, 64, 224, 224] 36,928
BatchNorm2d-5 [-1, 64, 224, 224] 128
ReLU-6 [-1, 64, 224, 224] 0
MaxPool2d-7 [-1, 64, 112, 112] 0
Conv2d-8 [-1, 128, 112, 112] 73,856
BatchNorm2d-9 [-1, 128, 112, 112] 256
ReLU-10 [-1, 128, 112, 112] 0
Conv2d-11 [-1, 128, 112, 112] 147,584
BatchNorm2d-12 [-1, 128, 112, 112] 256
ReLU-13 [-1, 128, 112, 112] 0
MaxPool2d-14 [-1, 128, 56, 56] 0
Conv2d-15 [-1, 256, 56, 56] 295,168
BatchNorm2d-16 [-1, 256, 56, 56] 512
ReLU-17 [-1, 256, 56, 56] 0
Conv2d-18 [-1, 256, 56, 56] 590,080
BatchNorm2d-19 [-1, 256, 56, 56] 512
ReLU-20 [-1, 256, 56, 56] 0
Conv2d-21 [-1, 256, 56, 56] 590,080
BatchNorm2d-22 [-1, 256, 56, 56] 512
ReLU-23 [-1, 256, 56, 56] 0
MaxPool2d-24 [-1, 256, 28, 28] 0
Conv2d-25 [-1, 512, 28, 28] 1,180,160
BatchNorm2d-26 [-1, 512, 28, 28] 1,024
ReLU-27 [-1, 512, 28, 28] 0
Conv2d-28 [-1, 512, 28, 28] 2,359,808
BatchNorm2d-29 [-1, 512, 28, 28] 1,024
ReLU-30 [-1, 512, 28, 28] 0
Conv2d-31 [-1, 512, 28, 28] 2,359,808
BatchNorm2d-32 [-1, 512, 28, 28] 1,024
ReLU-33 [-1, 512, 28, 28] 0
MaxPool2d-34 [-1, 512, 14, 14] 0
Conv2d-35 [-1, 512, 14, 14] 2,359,808
BatchNorm2d-36 [-1, 512, 14, 14] 1,024
ReLU-37 [-1, 512, 14, 14] 0
Conv2d-38 [-1, 512, 14, 14] 2,359,808
BatchNorm2d-39 [-1, 512, 14, 14] 1,024
ReLU-40 [-1, 512, 14, 14] 0
Conv2d-41 [-1, 512, 14, 14] 2,359,808
BatchNorm2d-42 [-1, 512, 14, 14] 1,024
ReLU-43 [-1, 512, 14, 14] 0
MaxPool2d-44 [-1, 512, 7, 7] 0
Dropout-45 [-1, 25088] 0
Linear-46 [-1, 4096] 102,764,544
ReLU-47 [-1, 4096] 0
Dropout-48 [-1, 4096] 0
Linear-49 [-1, 4096] 16,781,312
ReLU-50 [-1, 4096] 0
Linear-51 [-1, 17] 69,649
================================================================
Total params: 134,338,641
Trainable params: 134,338,641
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 322.10
Params size (MB): 512.46
Estimated Total Size (MB): 835.14
----------------------------------------------------------------
三、训练模型
1. 编写训练函数
# 训练循环
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集的大小
num_batches = len(dataloader) # 批次数目, (size/batch_size，向上取整)
train_loss, train_acc = 0, 0 # 初始化训练损失和正确率
for X, y in dataloader: # 获取图片及其标签
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model(X) # 网络输出
loss = loss_fn(pred, y) # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
# 反向传播
optimizer.zero_grad() # grad属性归零
loss.backward() # 反向传播
optimizer.step() # 每一步自动更新
# 记录acc与loss
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss += loss.item()
train_acc /= size
train_loss /= num_batches
return train_acc, train_loss
2. 编写测试函数
def test (dataloader, model, loss_fn):
size = len(dataloader.dataset) # 测试集的大小
num_batches = len(dataloader) # 批次数目, (size/batch_size，向上取整)
test_loss, test_acc = 0, 0
# 当不进行训练时，停止梯度更新，节省计算内存消耗
with torch.no_grad():
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算loss
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
test_loss += loss.item()
test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
3. 设置动态学习率
- 自定义动态学习率
# def adjust_learning_rate(optimizer, epoch, start_lr):
# # 每 2 个epoch衰减到原来的 0.98
# lr = start_lr * (0.92 ** (epoch // 2))
# for param_group in optimizer.param_groups:
# param_group['lr'] = lr
learn_rate = 1e-3 # 初始学习率
# optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
- 调用官方动态学习率
learn_rate = 1e-3
# 调用官方动态学习率接口时使用
lambda1 = lambda epoch: 0.92 ** (epoch // 4)
#optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
optimizer = torch.optim.Adam(model.parameters(), lr=learn_rate)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1) #选定调整方法
4. 正式训练
- 使用PyTorch 内置 VGG-16模型
import copy
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
epochs = 40
train_loss = []
train_acc = []
test_loss = []
test_acc = []
best_acc = 0 # 设置一个最佳准确率，作为最佳模型的判别指标
for epoch in range(epochs):
# 更新学习率（使用自定义学习率时使用）
# adjust_learning_rate(optimizer, epoch, learn_rate)
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）
model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
# 保存最佳模型到 best_model
if epoch_test_acc > best_acc:
best_acc = epoch_test_acc
best_model = copy.deepcopy(model)
train_acc.append(epoch_train_acc)
train_loss.append(epoch_train_loss)
test_acc.append(epoch_test_acc)
test_loss.append(epoch_test_loss)
# 获取当前的学习率
lr = optimizer.state_dict()['param_groups'][0]['lr']
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss,
epoch_test_acc*100, epoch_test_loss, lr))
# 保存最佳模型到文件中
PATH = './best_model.pth' # 保存的参数文件名
torch.save(best_model.state_dict(), PATH)
print('Done')
Epoch: 1, Train_acc:12.6%, Train_loss:2.813, Test_acc:25.3%, Test_loss:2.331, Lr:1.00E-03
- 使用自建VGG-16模型
import copy
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
epochs = 40
vgg16_train_loss = []
vgg16_train_acc = []
vgg16_test_loss = []
vgg16_test_acc = []
best_acc = 0 # 设置一个最佳准确率，作为最佳模型的判别指标
for epoch in range(epochs):
# 更新学习率（使用自定义学习率时使用）
# adjust_learning_rate(optimizer, epoch, learn_rate)
vgg16_model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, vgg16_model, loss_fn, optimizer)
scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）
vgg16_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, vgg16_model, loss_fn)
# 保存最佳模型到 best_model
if epoch_test_acc > best_acc:
best_acc = epoch_test_acc
vgg16_best_model = copy.deepcopy(vgg16_model)
vgg16_train_acc.append(epoch_train_acc)
vgg16_train_loss.append(epoch_train_loss)
vgg16_test_acc.append(epoch_test_acc)
vgg16_test_loss.append(epoch_test_loss)
# 获取当前的学习率
lr = optimizer.state_dict()['param_groups'][0]['lr']
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss,
epoch_test_acc*100, epoch_test_loss, lr))
# 保存最佳模型到文件中
PATH = './vgg16_best_model.pth' # 保存的参数文件名
torch.save(vgg16_best_model.state_dict(), PATH)
print('Done')
四、结果可视化
1. Loss 与 Accuracy 图
- 内置 VGG-16 模型
import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore") #忽略警告信息
#plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
plt.rcParams['figure.dpi'] = 100 #分辨率
epochs_range = range(epochs)
plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()
- 自建 VGG-16 模型
import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore") #忽略警告信息
#plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
plt.rcParams['figure.dpi'] = 100 #分辨率
epochs_range = range(epochs)
plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, vgg16_train_acc, label='Training Accuracy')
plt.plot(epochs_range, vgg16_test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1, 2, 2)
plt.plot(epochs_range, vgg16_train_loss, label='Training Loss')
plt.plot(epochs_range, vgg16_test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()
2. 指定图片进行预测
from PIL import Image
classes = list(total_data.class_to_idx)
def predict_one_image(image_path, model, transform, classes):
test_img = Image.open(image_path).convert('RGB')
plt.imshow(test_img) # 展示预测的图片
test_img = transform(test_img)
img = test_img.to(device).unsqueeze(0)
model.eval()
output = model(img)
_,pred = torch.max(output,1)
pred_class = classes[pred]
print(f'预测结果是：{pred_class}')
# 预测训练集中的某张照片
predict_one_image(image_path='./small_data/Tom Hanks/100_b712e7ca.jpg',
model=model,
transform=train_transforms,
classes=classes)
预测结果是：Hugh Jackman
3. 模型评估
best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)
epoch_test_acc, epoch_test_loss
(0.24444444444444444, 2.323750692865123)
# 查看是否与我们记录的最高准确率一致
epoch_test_acc
0.85
五、总结
- 基本掌握了如何使用 PyTorch 自建模型。
- 使用自建 VGG-16 模型的时候，经常报错OutOfMemoryError；当缩小训练数据集为3个明星的时候，还是报错，自建VGG-16模型无法运行起来。
- 暂时还不知道如何优化模型的，使其达到60%的测试集准确率；网上查阅了资料显示可以通过: (1) 更改优化器，比如使用Adam优化器；(2) 调整模型内部全连接层参数；(3) 训练数据集的照片添加随机水平翻转。经过尝试后，模型的测试集准确率达到了60%。
