- 🍨 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:
- 自己搭建VGG-16网络框架
- 调用官方的VGG-16网络框架
- 如何查看模型的参数量以及相关指标
🍻 拔高:
- 验证集准确率达到100%
- 使用PPT画出VGG-16算法框架图(发论文需要这项技能!!)
🔎 探索:
- 在不影响准确率的前提下轻量化模型
● 目前VGG16的Total params是134,272,835
🏡 我的环境:
● 语言环境:Python3.8
● 编译器:Jupyter Lab
● 深度学习环境:Pytorch
○ torch == 2.2.2 + cu121
○ torchvision == 0.17.2 + cu121
一、前期准备
1.设置GPU
import torch
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms
from torchvision import transforms, datasets
import os, PIL, pathlib, warnings
warnings.filterwarnings('ignore')
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
device
device(type='cuda')
2.导入数据
import os, PIL, random, pathlib
data_dir = './data/P7/PotatoPlants/'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classNames = [str(path).split('\\')[3] for path in data_paths]
classNames
['Early_blight', 'healthy', 'Late_blight']
train_transforms = transforms.Compose([
transforms.Resize([224, 224]),
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
])
test_transforms = transforms.Compose([
transforms.Resize([224, 224]),
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
])
total_data = datasets.ImageFolder('./data/P7/PotatoPlants/', transform = train_transforms)
total_data
Dataset ImageFolder
Number of datapoints: 2152
Root location: ./data/P7/PotatoPlants/
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=True)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
total_data.class_to_idx
{'Early_blight': 0, 'Late_blight': 1, 'healthy': 2}
3.划分数据集
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 0x1cf705d7bb0>,
<torch.utils.data.dataset.Subset at 0x1cf70bd3100>)
batch_size = 32
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([32, 3, 224, 224])
Shape of y: torch.Size([32]) torch.int64
二、手动搭建VGG-16模型
1.搭建模型
import torch.nn.functional as F
class vgg16(nn.Module):
def __init__(self):
super(vgg16, self).__init__()
# 卷积块1
self.block1 = nn.Sequential(
nn.Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2))
)
# 卷积块2
self.block2 = nn.Sequential(
nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2)
)
# 卷积块3
self.block3 = nn.Sequential(
nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2)
)
# 卷积块4
self.block4 = nn.Sequential(
nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2)
)
# 卷积块5
self.block5 = nn.Sequential(
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),
nn.ReLU(),
nn.MaxPool2d(kernel_size=2, stride=2)
)
# 全连接网络层,用于分类
self.classifier = nn.Sequential(
nn.Linear(in_features=512*7*7, out_features=4096),
nn.ReLU(),
nn.Linear(in_features=4096, out_features=4096),
nn.ReLU(),
nn.Linear(in_features=4096, out_features=3)
)
def forward(self, x):
x = self.block1(x)
x = self.block2(x)
x = self.block3(x)
x = self.block4(x)
x = self.block5(x)
x = torch.flatten(x, start_dim = 1)
x = self.classifier(x)
return x
print('Using {} device'.format(device))
model = vgg16().to(device)
model
Using cuda device
vgg16(
(block1): Sequential(
(0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU()
(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU()
(4): MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=0, dilation=1, ceil_mode=False)
)
(block2): Sequential(
(0): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU()
(2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU()
(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(block3): Sequential(
(0): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU()
(2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU()
(4): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(5): ReLU()
(6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(block4): Sequential(
(0): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU()
(2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU()
(4): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(5): ReLU()
(6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(block5): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(1): ReLU()
(2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(3): ReLU()
(4): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(5): ReLU()
(6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
)
(classifier): Sequential(
(0): Linear(in_features=25088, out_features=4096, bias=True)
(1): ReLU()
(2): Linear(in_features=4096, out_features=4096, bias=True)
(3): ReLU()
(4): Linear(in_features=4096, out_features=3, bias=True)
)
)
2.查看模型详情
import torchsummary as summary
summary.summary(model, (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
Linear-32 [-1, 4096] 102,764,544
ReLU-33 [-1, 4096] 0
Linear-34 [-1, 4096] 16,781,312
ReLU-35 [-1, 4096] 0
Linear-36 [-1, 3] 12,291
================================================================
Total params: 134,272,835
Trainable params: 134,272,835
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 218.52
Params size (MB): 512.21
Estimated Total Size (MB): 731.30
----------------------------------------------------------------
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)
train_loss, train_acc = 0, 0
for X, y in dataloader:
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model(X)
loss = loss_fn(pred, y)
# 反向传播
optimizer.zero_grad()
loss.backward()
optimizer.step()
# 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
# 测试循环
def test(dataloader, model, loss_fn):
size = len(dataloader.dataset)
num_batches = len(dataloader)
test_loss, test_acc = 0, 0
with torch.no_grad(): # 不进行训练时,停止梯度更新
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算预测误差
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
# 记录acc与loss
test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_loss += loss.item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
3.正式训练
import copy
optimizer = torch.optim.Adam(model.parameters(), lr= 1e-4)
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):
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
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 = './model/best_model_P7.pth' # 保存的参数文件名
torch.save(model.state_dict(), PATH)
print('Done')
Epoch: 1, Train_acc:48.2%, Train_loss:0.924, Test_acc:48.5%, Test_loss:0.910, Lr:1.00E-04
Epoch: 2, Train_acc:46.0%, Train_loss:0.913, Test_acc:48.5%, Test_loss:0.915, Lr:1.00E-04
Epoch: 3, Train_acc:46.4%, Train_loss:0.910, Test_acc:44.1%, Test_loss:0.925, Lr:1.00E-04
Epoch: 4, Train_acc:47.4%, Train_loss:0.904, Test_acc:44.1%, Test_loss:0.916, Lr:1.00E-04
Epoch: 5, Train_acc:46.7%, Train_loss:0.905, Test_acc:48.5%, Test_loss:0.909, Lr:1.00E-04
Epoch: 6, Train_acc:48.3%, Train_loss:0.900, Test_acc:44.1%, Test_loss:0.941, Lr:1.00E-04
Epoch: 7, Train_acc:45.3%, Train_loss:0.905, Test_acc:48.5%, Test_loss:0.905, Lr:1.00E-04
Epoch: 8, Train_acc:47.9%, Train_loss:0.899, Test_acc:44.1%, Test_loss:0.911, Lr:1.00E-04
Epoch: 9, Train_acc:45.7%, Train_loss:0.904, Test_acc:44.1%, Test_loss:0.914, Lr:1.00E-04
Epoch:10, Train_acc:46.7%, Train_loss:0.903, Test_acc:44.1%, Test_loss:0.922, Lr:1.00E-04
......
Epoch:31, Train_acc:98.3%, Train_loss:0.039, Test_acc:95.4%, Test_loss:0.232, Lr:1.00E-04
Epoch:32, Train_acc:99.4%, Train_loss:0.017, Test_acc:94.4%, Test_loss:0.216, Lr:1.00E-04
Epoch:33, Train_acc:99.5%, Train_loss:0.012, Test_acc:96.1%, Test_loss:0.465, Lr:1.00E-04
Epoch:34, Train_acc:98.8%, Train_loss:0.030, Test_acc:96.3%, Test_loss:0.162, Lr:1.00E-04
Epoch:35, Train_acc:99.4%, Train_loss:0.017, Test_acc:96.3%, Test_loss:0.327, Lr:1.00E-04
Epoch:36, Train_acc:99.1%, Train_loss:0.026, Test_acc:95.8%, Test_loss:0.226, Lr:1.00E-04
Epoch:37, Train_acc:98.9%, Train_loss:0.024, Test_acc:95.6%, Test_loss:0.292, Lr:1.00E-04
Epoch:38, Train_acc:95.1%, Train_loss:0.124, Test_acc:95.4%, Test_loss:0.164, Lr:1.00E-04
Epoch:39, Train_acc:98.8%, Train_loss:0.031, Test_acc:95.6%, Test_loss:0.261, Lr:1.00E-04
Epoch:40, Train_acc:99.1%, Train_loss:0.024, Test_acc:95.8%, Test_loss:0.205, Lr:1.00E-04
Done
四、结果可视化
1.Loss与Accuracy图
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()
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='./data/P7/PotatoPlants/Early_blight/1defd277-6394-4049-895d-470f1e27e189___RS_Early.B 8931.JPG',
model=model,
transform=train_transforms,
classes=classes)
预测结果是:Early_blight
3.模型评估
best_model.eval()
epoch_best_acc, epoch_test_loaa = test(test_dl, best_model, loss_fn)
epoch_test_acc, epoch_test_loss
(0.9582366589327146, 0.20503126973718672)
# 查看是否与我们记录的最高准确率一致
epoch_test_acc
0.9582366589327146
五、总结
用符合要求的数据训练的模型表现效果会更好,如果能在预训练模型的基础上再用自己的数据进行训练可能会表现更好。(自己留个坑先)