- 🍨 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 | 接辅导、项目定制
- 🚀 文章来源：K同学的学习圈子
目录
我的环境
- 语言环境：python3.8.18
- 编译器：jupyter notebook
- 深度学习环境：torch==2.0.1+cu118，torchvision==0.15.2+cu118
一、 前期准备
1. 设置GPU
如果设备上支持GPU就使用GPU,否则使用CPU
import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os,PIL,pathlib
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
devicedevice(type='cuda')2. 导入数据
import os,PIL,random,pathlib
data_dir = 'E:\\pytorch练习\\训练营\\P7\\咖啡豆数据集-K同学啊'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[5] for path in data_paths]
classeNames['Dark', 'Green', 'Light', 'Medium']# 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。
])
test_transform = transforms.Compose([
transforms.Resize([224, 224]), # 将输入图片resize成统一尺寸
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("E:\\pytorch练习\\训练营\\P7\\咖啡豆数据集-K同学啊",transform=train_transforms)
total_dataDataset ImageFolder
Number of datapoints: 1200
Root location: E:\pytorch练习\训练营\P7\咖啡豆数据集-K同学啊
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=warn)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)total_data.class_to_idx{'Dark': 0, 'Green': 1, 'Light': 2, 'Medium': 3}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 0x2c7957f97e0>,
<torch.utils.data.dataset.Subset at 0x2c7957f96f0>)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)
breakShape 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
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），分别用
fcX与predictions表示 - 5个池化层（Pool layer），分别用
blockX_pool表示
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
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, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2))
)
# 卷积块3
self.block3 = nn.Sequential(
nn.Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2))
)
# 卷积块4
self.block4 = nn.Sequential(
nn.Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2))
)
# 卷积块5
self.block5 = nn.Sequential(
nn.Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 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=4)
)
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
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
model = vgg16().to(device)
modelUsing 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, 2), stride=(2, 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, 2), stride=(2, 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, 2), stride=(2, 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, 2), stride=(2, 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=4, bias=True)
)
)2. 查看模型详情
# 统计模型参数量以及其他指标
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, 4] 16,388
================================================================
Total params: 134,276,932
Trainable params: 134,276,932
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 218.52
Params size (MB): 512.23
Estimated Total Size (MB): 731.32
----------------------------------------------------------------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_loss2. 编写测试函数
测试函数和训练函数大致相同，但是由于不进行梯度下降对网络权重进行更新，所以不需要传入优化器
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_loss3. 正式训练
model.train()、model.eval()训练营往期文章中有详细的介绍。
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 = './best_model.pth' # 保存的参数文件名
torch.save(model.state_dict(), PATH)
print('Done')Epoch: 1, Train_acc:22.1%, Train_loss:1.388, Test_acc:22.5%, Test_loss:1.387, Lr:1.00E-04
Epoch: 2, Train_acc:28.3%, Train_loss:1.350, Test_acc:65.4%, Test_loss:0.840, Lr:1.00E-04
Epoch: 3, Train_acc:59.0%, Train_loss:0.845, Test_acc:74.2%, Test_loss:0.615, Lr:1.00E-04
Epoch: 4, Train_acc:70.4%, Train_loss:0.583, Test_acc:80.8%, Test_loss:0.512, Lr:1.00E-04
Epoch: 5, Train_acc:75.8%, Train_loss:0.477, Test_acc:82.9%, Test_loss:0.353, Lr:1.00E-04
Epoch: 6, Train_acc:81.4%, Train_loss:0.399, Test_acc:84.2%, Test_loss:0.324, Lr:1.00E-04
Epoch: 7, Train_acc:86.9%, Train_loss:0.320, Test_acc:94.6%, Test_loss:0.198, Lr:1.00E-04
Epoch: 8, Train_acc:89.2%, Train_loss:0.282, Test_acc:87.9%, Test_loss:0.257, Lr:1.00E-04
Epoch: 9, Train_acc:92.6%, Train_loss:0.229, Test_acc:92.5%, Test_loss:0.229, Lr:1.00E-04
Epoch:10, Train_acc:90.4%, Train_loss:0.299, Test_acc:92.1%, Test_loss:0.310, Lr:1.00E-04
Epoch:11, Train_acc:91.5%, Train_loss:0.251, Test_acc:91.2%, Test_loss:0.237, Lr:1.00E-04
Epoch:12, Train_acc:94.7%, Train_loss:0.158, Test_acc:93.8%, Test_loss:0.149, Lr:1.00E-04
Epoch:13, Train_acc:94.0%, Train_loss:0.149, Test_acc:90.4%, Test_loss:0.201, Lr:1.00E-04
Epoch:14, Train_acc:94.7%, Train_loss:0.173, Test_acc:94.6%, Test_loss:0.183, Lr:1.00E-04
Epoch:15, Train_acc:94.1%, Train_loss:0.170, Test_acc:97.1%, Test_loss:0.069, Lr:1.00E-04
Epoch:16, Train_acc:96.8%, Train_loss:0.098, Test_acc:90.8%, Test_loss:0.277, Lr:1.00E-04
Epoch:17, Train_acc:95.5%, Train_loss:0.136, Test_acc:96.2%, Test_loss:0.072, Lr:1.00E-04
Epoch:18, Train_acc:95.6%, Train_loss:0.127, Test_acc:95.8%, Test_loss:0.127, Lr:1.00E-04
Epoch:19, Train_acc:96.8%, Train_loss:0.100, Test_acc:98.3%, Test_loss:0.043, Lr:1.00E-04
Epoch:20, Train_acc:96.7%, Train_loss:0.082, Test_acc:97.1%, Test_loss:0.066, Lr:1.00E-04
Epoch:21, Train_acc:98.3%, Train_loss:0.053, Test_acc:98.3%, Test_loss:0.052, Lr:1.00E-04
Epoch:22, Train_acc:97.9%, Train_loss:0.056, Test_acc:98.3%, Test_loss:0.052, Lr:1.00E-04
Epoch:23, Train_acc:97.4%, Train_loss:0.081, Test_acc:99.2%, Test_loss:0.051, Lr:1.00E-04
Epoch:24, Train_acc:98.2%, Train_loss:0.055, Test_acc:97.1%, Test_loss:0.063, Lr:1.00E-04
Epoch:25, Train_acc:97.1%, Train_loss:0.086, Test_acc:91.7%, Test_loss:0.201, Lr:1.00E-04
Epoch:26, Train_acc:95.9%, Train_loss:0.128, Test_acc:97.1%, Test_loss:0.105, Lr:1.00E-04
Epoch:27, Train_acc:98.2%, Train_loss:0.054, Test_acc:99.2%, Test_loss:0.036, Lr:1.00E-04
Epoch:28, Train_acc:97.8%, Train_loss:0.056, Test_acc:96.7%, Test_loss:0.099, Lr:1.00E-04
Epoch:29, Train_acc:98.0%, Train_loss:0.059, Test_acc:98.8%, Test_loss:0.057, Lr:1.00E-04
Epoch:30, Train_acc:98.8%, Train_loss:0.036, Test_acc:97.1%, Test_loss:0.050, Lr:1.00E-04
Epoch:31, Train_acc:97.2%, Train_loss:0.061, Test_acc:96.7%, Test_loss:0.094, Lr:1.00E-04
Epoch:32, Train_acc:99.6%, Train_loss:0.015, Test_acc:98.8%, Test_loss:0.045, Lr:1.00E-04
Epoch:33, Train_acc:97.0%, Train_loss:0.096, Test_acc:98.3%, Test_loss:0.090, Lr:1.00E-04
Epoch:34, Train_acc:98.1%, Train_loss:0.058, Test_acc:97.9%, Test_loss:0.050, Lr:1.00E-04
Epoch:35, Train_acc:99.0%, Train_loss:0.030, Test_acc:98.8%, Test_loss:0.050, Lr:1.00E-04
Epoch:36, Train_acc:98.3%, Train_loss:0.047, Test_acc:98.3%, Test_loss:0.054, Lr:1.00E-04
Epoch:37, Train_acc:99.3%, Train_loss:0.022, Test_acc:97.9%, Test_loss:0.100, Lr:1.00E-04
Epoch:38, Train_acc:99.2%, Train_loss:0.017, Test_acc:99.2%, Test_loss:0.034, Lr:1.00E-04
Epoch:39, Train_acc:99.1%, Train_loss:0.019, Test_acc:99.2%, Test_loss:0.085, Lr:1.00E-04
Epoch:40, Train_acc:99.8%, Train_loss:0.005, Test_acc:97.9%, Test_loss:0.086, Lr:1.00E-04四、 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='E:\\pytorch练习\\训练营\\P7\\咖啡豆数据集-K同学啊\\Dark\\dark (1).png',
model=model,
transform=train_transforms,
classes=classes)
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(0.9916666666666667, 0.035762640996836126)# 查看是否与我们记录的最高准确率一致
epoch_test_acc0.9916666666666667五、个人总结
本周学习了VGG-16模型的手动实现，通过动手复现更加理解了该模型的底层原理，也对我的专业方向有了一个新的启发，可以将交通流数据的图看做文章中图片的格式，将其transform，最后的分类问题转换成回归问题。
扫一扫
2万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
