一,前言(引用)
>- **🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/Z9yL_wt7L8aPOr9Lqb1K3w) 中的学习记录博客**
>- **🍖 原作者:[K同学啊](https://mtyjkh.blog.csdn.net/)**
二,训练准备
1.训练环境
- Framework: Pytorch, matplotlib,numpy
- Compiler:Jupyter Lab
- Cpu: AMD Ryzen 5600H
2.数据集
4种天气图片jpg,分别存储在['cloudy', 'rain', 'shine', 'sunrise']标签下,图片总数为: 1125
Cloudy数据集示例:
# 指定图像文件夹路径
image_folder = 'E:\Basic Training 365\P3 Wetter Diagnose\weather_photos/cloudy/'
# 获取文件夹中的所有图像文件
image_files = [f for f in os.listdir(image_folder) if f.endswith((".jpg", ".png", ".jpeg"))]
# 创建Matplotlib图像
fig, axes = plt.subplots(3, 8, figsize=(16, 6))
# 使用列表推导式加载和显示图像
for ax, img_file in zip(axes.flat, image_files):
img_path = os.path.join(image_folder, img_file)
img = Image.open(img_path)
ax.imshow(img)
ax.axis('off')
# 显示图像
plt.tight_layout()
plt.show()
三,训练过程
初始化:
import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os,PIL,pathlib,random
import matplotlib.pyplot as plt
from PIL import Image
import warnings
数据集:
data_dir = 'E:\Basic Training 365\P3 Wetter Diagnose\weather_photos'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[1] for path in data_paths]
total_datadir = 'E:/Basic Training 365/P3 Wetter Diagnose/weather_photos'
train_transforms = 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(total_datadir,transform=train_transforms)
- (<torch.utils.data.dataset.Subset at 0x1ea49ffc6a0>,
<torch.utils.data.dataset.Subset at 0x1ea49ffe9b0>) - Dataset ImageFolder Number of datapoints: 1125 Root location: E:/Basic Training 365/P3 Wetter Diagnose/weather_photos 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]) )
训练配置:
torch.utils.data.random_split()
方法进行数据集划分。该方法将总体数据total_data按照指定的大小比例([train_size, test_size])随机划分为训练集和测试集,并将划分结果分别赋值给train_dataset和test_dataset两个变量。
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])
print(train_size,test_size)
(900, 225)
torch.utils.data.DataLoader():
- dataset(必需参数):这是你的数据集对象,通常是
torch.utils.data.Dataset
的子类,它包含了你的数据样本。 - batch_size(可选参数):指定每个小批次中包含的样本数。默认值为 1。
- shuffle(可选参数):如果设置为
True
,则在每个 epoch 开始时对数据进行洗牌,以随机打乱样本的顺序。这对于训练数据的随机性很重要,以避免模型学习到数据的顺序性。默认值为False
。 - num_workers(可选参数):用于数据加载的子进程数量。通常,将其设置为大于 0 的值可以加快数据加载速度,特别是当数据集很大时。默认值为 0,表示在主进程中加载数据。
- pin_memory(可选参数):如果设置为
True
,则数据加载到 GPU 时会将数据存储在 CUDA 的锁页内存中,这可以加速数据传输到 GPU。默认值为False
。 - drop_last(可选参数):如果设置为
True
,则在最后一个小批次可能包含样本数小于batch_size
时,丢弃该小批次。这在某些情况下很有用,以确保所有小批次具有相同的大小。默认值为False
。 - timeout(可选参数):如果设置为正整数,它定义了每个子进程在等待数据加载器传递数据时的超时时间(以秒为单位)。这可以用于避免子进程卡住的情况。默认值为 0,表示没有超时限制。
- worker_init_fn(可选参数):一个可选的函数,用于初始化每个子进程的状态。这对于设置每个子进程的随机种子或其他初始化操作很有用。
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
设计CNN训练模型:
-
torch.nn.Conv2d()函数:
torch.nn.Conv2d(in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True, padding_mode='zeros', device=None, dtype=None)
第一个参数(in_channels)是输入的channel数量
第二个参数(out_channels)是输出的channel数量
第三个参数(kernel_size)是卷积核大小
第四个参数(stride)是步长,默认为1
第五个参数(padding)是填充大小,默认为0
-
torch.nn.MaxPool2d()
torch.nn.MaxPool2d(kernel_size, stride=None, padding=0, dilation=1, return_indices=False, ceil_mode=False)
in_channels ( int ) – 输入图像中的通道数
out_channels ( int ) – 卷积产生的通道数
kernel_size ( int or tuple ) – 卷积核的大小
stride ( int or tuple , optional ) -- 卷积的步幅。默认值:1
padding ( int , tuple或str , optional ) – 添加到输入的所有四个边的填充。默认值:0
padding_mode (字符串,可选) – 'zeros', 'reflect', 'replicate'或'circular'. 默认:'zeros'
import torch.nn.functional as F
class Network_bn(nn.Module):
def __init__(self):
super(Network_bn, self).__init__()
self.conv1 = nn.Conv2d(in_channels=3, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn1 = nn.BatchNorm2d(12) #Batch Normalization
self.conv2 = nn.Conv2d(in_channels=12, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn2 = nn.BatchNorm2d(12) #正则化,稳定输出减少偏移
self.pool1 = nn.MaxPool2d(2,2)
self.conv4 = nn.Conv2d(in_channels=12, out_channels=24, kernel_size=5, stride=1, padding=0)
self.bn4 = nn.BatchNorm2d(24)
self.conv5 = nn.Conv2d(in_channels=24, out_channels=24, kernel_size=5, stride=1, padding=0)
self.bn5 = nn.BatchNorm2d(24)
self.pool2 = nn.MaxPool2d(2,2)
self.fc1 = nn.Linear(24*50*50, len(classeNames))
def forward(self, x):
x = F.relu(self.bn1(self.conv1(x)))
x = F.relu(self.bn2(self.conv2(x)))
x = self.pool1(x)
x = F.relu(self.bn4(self.conv4(x)))
x = F.relu(self.bn5(self.conv5(x)))
x = self.pool2(x)
x = x.view(-1, 24*50*50)
x = self.fc1(x)
return x
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
model = Network_bn().to(device)
得到模型结构:
Network_bn(
(conv1): Conv2d(3, 12, kernel_size=(5, 5), stride=(1, 1))
(bn1): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(12, 12, kernel_size=(5, 5), stride=(1, 1))
(bn2): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(pool1): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(conv4): Conv2d(12, 24, kernel_size=(5, 5), stride=(1, 1))
(bn4): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv5): Conv2d(24, 24, kernel_size=(5, 5), stride=(1, 1))
(bn5): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(pool2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(fc1): Linear(in_features=60000, out_features=4, bias=True)
)
符合设计大小:
3, 224, 224
(输入数据)
-> 12, 220, 220
(经过卷积层1)
-> 12, 216, 216
(经过卷积层2)-> 12, 108, 108
(经过池化层1)
-> 24, 104, 104
(经过卷积层3)
-> 24, 100, 100
(经过池化层4)-> 24, 50, 50
(经过池化层2)
-> 60000
创建损失函数:
loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-4 # 学习率
opt = torch.optim.SGD(model.parameters(),lr=learn_rate)
训练循环设计:
# 训练循环
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集的大小,一共60000张图片
num_batches = len(dataloader) # 批次数目,1875(60000/32)
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
测试函数设计:
def test (dataloader, model, loss_fn):
size = len(dataloader.dataset) # 测试集的大小,一共10000张图片
num_batches = len(dataloader) # 批次数目,313(10000/32=312.5,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
训练过程:
epochs = 20
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)
model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
train_acc.append(epoch_train_acc)
train_loss.append(epoch_train_loss)
test_acc.append(epoch_test_acc)
test_loss.append(epoch_test_loss)
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%,Test_loss:{:.3f}')
print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss))
print('Done')
Epoch: 1, Train_acc:64.1%, Train_loss:0.990, Test_acc:76.9%,Test_loss:0.871
Epoch: 2, Train_acc:80.6%, Train_loss:0.644, Test_acc:78.2%,Test_loss:0.557
Epoch: 3, Train_acc:84.6%, Train_loss:0.527, Test_acc:88.4%,Test_loss:0.381
Epoch: 4, Train_acc:87.3%, Train_loss:0.445, Test_acc:88.0%,Test_loss:0.357
Epoch: 5, Train_acc:89.2%, Train_loss:0.409, Test_acc:88.0%,Test_loss:0.368
Epoch: 6, Train_acc:90.2%, Train_loss:0.343, Test_acc:89.3%,Test_loss:0.298
Epoch: 7, Train_acc:90.3%, Train_loss:0.328, Test_acc:90.2%,Test_loss:0.276
Epoch: 8, Train_acc:91.8%, Train_loss:0.318, Test_acc:87.6%,Test_loss:0.440
Epoch: 9, Train_acc:92.0%, Train_loss:0.277, Test_acc:90.2%,Test_loss:0.259
Epoch:10, Train_acc:91.8%, Train_loss:0.270, Test_acc:87.1%,Test_loss:0.289
Epoch:11, Train_acc:91.2%, Train_loss:0.290, Test_acc:85.3%,Test_loss:0.329
Epoch:12, Train_acc:92.8%, Train_loss:0.261, Test_acc:92.0%,Test_loss:0.241
Epoch:13, Train_acc:94.1%, Train_loss:0.233, Test_acc:92.0%,Test_loss:0.260
Epoch:14, Train_acc:92.6%, Train_loss:0.249, Test_acc:92.0%,Test_loss:0.230
Epoch:15, Train_acc:92.8%, Train_loss:0.232, Test_acc:91.1%,Test_loss:0.245
Epoch:16, Train_acc:95.0%, Train_loss:0.198, Test_acc:91.6%,Test_loss:0.224
Epoch:17, Train_acc:95.2%, Train_loss:0.196, Test_acc:91.1%,Test_loss:0.227
Epoch:18, Train_acc:94.6%, Train_loss:0.227, Test_acc:87.1%,Test_loss:0.318
Epoch:19, Train_acc:93.9%, Train_loss:0.206, Test_acc:92.9%,Test_loss:0.229
Epoch:20, Train_acc:94.7%, Train_loss:0.208, Test_acc:92.4%,Test_loss:0.224
Done
输出测试结果:
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()
四、总结
四、总结
在本次学习中,我们实现了一个基于PyTorch框架的天气图像分类模型。通过完整的深度学习项目实践,深入了解了从数据准备、模型设计到训练和评估的全流程。以下是具体的学习收获:
-
训练准备与数据集处理:
- 使用PyTorch框架及其相关库(如
torchvision
、matplotlib
)来进行图像分类任务。 - 数据集由四类天气图片组成,分别为
cloudy
、rain
、shine
和sunrise
,共有1125张图片。通过ImageFolder
方法组织这些图片,并对图片进行预处理(如调整尺寸、归一化处理),为模型的训练提供了清晰而规范的数据输入格式。
- 使用PyTorch框架及其相关库(如
-
模型设计与实现:
- 构建了一个卷积神经网络(CNN)模型
Network_bn
,使用了多层卷积和池化操作来提取图像的特征。每个卷积层之后都进行了批量归一化(Batch Normalization),以稳定模型的训练过程。 - 模型结构包括了多层卷积层、池化层和一个全连接层,分别负责特征提取和最终的分类决策。通过设计网络的前向传播过程,学习了如何将原始输入转换为输出标签。
- 构建了一个卷积神经网络(CNN)模型
-
模型训练与评估:
- 使用了交叉熵损失函数(CrossEntropyLoss)和随机梯度下降优化器(SGD),通过20个周期的训练逐步优化模型参数。数据集被分为训练集和测试集,分别占80%和20%。
- 通过设计训练和测试函数,分别计算模型在训练和测试过程中的损失和准确度。绘制了训练和测试的准确率及损失曲线,展示了模型性能的变化趋势。
- 最终结果显示,模型在训练集上的准确率达到94.7%,在测试集上的准确率为92.4%,表明模型具有较好的泛化能力和较高的分类准确率。