🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/rbOOmire8OocQ90QM78DRA) 中的学习记录博客
🍖 原作者:[K同学啊 | 接辅导、项目定制](https://mtyjkh.blog.csdn.net/)
一、前期准备
1. 导入必要的库
import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os, PIL, pathlib, random
device = torch.device("cuda" if torch.cuda.is_available() else "cpu") #print(type = "cuda")
2. 导入数据
创建pathlib.path路径对象,读取该文件夹下所有文件路径,并获取文件名
data_dir = './data/'
data_dir = pathlib.Path(data_dir) #创建路径对象
data_paths = list(data_dir.glob('*')) #获取文件夹下所有文件路径
classeNames = [str(path).split("\\")[1] for path in data_paths] #将路径转化为字符串并切割出文件名(不包括拓展名)
classeNames
## data_paths = list(pathlib.Path('./data.').glob('*')) #获取data文件夹下所有文件的路径
## classeNames = [str(path).split("\\")[1] for path in data_paths] #获取文件名(不包括拓展名
PIL
(Python Imaging Library)中的Image
模块用于处理图像。
import matplotlib.pyplot as plt
from PIL import Image
image_folder = './data/cloudy/' #图像文件所在路径
# 获取文件夹中所有图像文件
image_files = [f for f in os.listdir(image_folder) if f.endswith((".jpg", "jpeg", ".png"))]
# 创建matplotlib图像
fig, axes = plt.subplots(3, 8, fig size = (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) #拼接每个图像文件的路径
img = Image.open(img_path)
ax.imshow(img)
ax.axis('off')
# 显示图像
plt.tight_layout() #调整子图间距,紧凑排列
plt.show()
pathlib.Path 和 os.listdir 用法:
from pathlib import Path # 列出当前目录下的所有文件 for item in Path('.').iterdir(): print(item) # 获取当前目录下所有txt文件的路径 txt_files = Path('.').glob('*.txt') for file in txt_files: print(file) # 检查'example.txt'文件是否存在 if (Path('.') / 'example.txt').exists(): print("File exists!") else: print("File does not exist.")
import os # 列出当前目录下的所有文件和目录 items = os.listdir('.') for item in items: print(item) # 获取当前目录下所有txt文件的路径 txt_files = [file for file in os.listdir('.') if file.endswith('.txt')] for file in txt_files: print(file) # 检查'example.txt'文件是否存在 if 'example.txt' in os.listdir('.'): print("File exists!") else: print("File does not exist.")
创建图像数据集并对PIL Image做转换:
将输入图片调整为统一尺寸(224x224像素)
转化为tensor,转化为标准正态分布
total_datadir = './data/'
# 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.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-->转换为标准正态分布(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)
total_data
输出:
Dataset ImageFolder
Number of datapoints: 1125
Root location: ./data/
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=None)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
3. 划分数据集
按照80%和20%的比例随机划分为train_dataset
和test_dataset;
torch.utils.data.random_split()
函数将total_data
按照给定的[train_size, test_size]
比例进行随机划分:
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
输出:
(<torch.utils.data.dataset.Subset at 0x1cd91e01ee0>,
<torch.utils.data.dataset.Subset at 0x1cd91e01f70>)
torch.utils.data.DataLoader()加载数据集:
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)
打印测试集形状(只打印一次):
N--批次数,C-通道数,H--高度,W--宽度
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
输出:
Shape of X [N, C, H, W]: torch.Size([32, 3, 224, 224])
Shape of y: torch.Size([32]) torch.int64
torch.utils.data.DataLoader参数解释:
dataset
:需要加载的数据集对象,通常是torchvision.datasets
或自定义的数据集类。batch_size
:每个批次的数据量大小。shuffle
:是否在每个训练周期开始时打乱数据顺序,默认为False
。num_workers
:用于数据加载的工作进程数,默认为0,表示数据将在主进程中加载。增加num_workers
可以加速数据加载。pin_memory
:是否将数据存储在CUDA固定内存中,以加快GPU数据传输速度,默认为False
。drop_last
:当数据集大小不能被batch_size
整除时,是否丢弃最后一个不完整的批次,默认为False
。timeout
(可选):从工作进程获取数据的超时时间(秒),默认为0。worker_init_fn
(可选):用于初始化工作进程的函数,默认为None
。collate_fn
(可选):将一个batch的数据进行整理的函数,默认为None
。
二、构建简单的CNN网络
继承nn.Module
创建自定义的神经网络层和模型
import torch.nn.functional as F
class Network_bn(nn.Module):
def __init__(self):
super(Network_bn, self).__init__()
"""
nn.Conv2d()函数:
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
*************************************************
nn.BatchNorm2d()(实现二维批量归一化操作):
num_features:是输入张量的特征数量,通常对应于输入图像的通道数C
*************************************************
nn.Linear():
通道*高*宽,输出特征维度等于类别数
"""
self.conv1 = nn.Conv2d(in_channels=3, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn1 = nn.BatchNorm2d(12)
self.conv2 = nn.Conv2d(in_channels=12, out_channels=12, kernel_size=5, stride=1, padding=0)
self.bn2 = nn.BatchNorm2d(12)
self.pool = 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.fc1 = nn.Linear(24*50*50, len(classeNames))
def forward(self, x):
x = F.relu(self.bn1(self.conv1(x))) #卷积-批量归一化-激活
x = F.relu(self.bn2(self.conv2(x))) #卷积-批量归一化-激活
x = self.pool(x) #池化
x = F.relu(self.bn4(self.conv4(x))) #卷积-批量归一化-激活
x = F.relu(self.bn5(self.conv5(x))) #卷积-批量归一化-激活
x = self.pool(x) #池化
x = x.view(-1, 24*50*50)
x = self.fc1(x) #全连接
return x
## 检查是否支持CUDA,选并将模型移动到相应的设备上,返回模型对象:
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
model = Network_bn().to(device)
model
输出:
Using cuda device
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)
(pool): 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)
(fc1): Linear(in_features=60000, out_features=4, bias=True)
)
三、训练模型
1. 设置超参数
损失函数、学习率、随即梯度下降优化器
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
learn_rate = 1e-4 # 学习率
opt = torch.optim.SGD(model.parameters(),lr=learn_rate)
2. 编写训练函数
1. optimizer.zero_grad()
用于将优化器中的所有梯度(即模型参数的梯度)清零。在每次迭代参数之前,都需要调用这个方法来清除之前的梯度,以确保只计算当前迭代的梯度;
2. loss.backward()
负责计算梯度;
3. optimizer.step()
则使用这些梯度来更新模型参数。
通过len(dataloader.dataset)
获取训练集的大小;
通过len(dataloader)
获取批次数目;
初始化训练损失(train_loss
)和正确率(train_acc
)为0;
使用for
循环遍历数据加载器中的每个批次。在每次迭代中,首先将输入数据(X
)和目标标签(y
)移动到设备(device
)上;
通过模型(model
)计算预测结果(pred
),并使用损失函数(loss_fn
)计算预测结果和真实标签之间的loss;
进行反向传播,首先调用优化器的zero_grad()
--loss.backward()
--optimizer.step();
在每次迭代中,记录正确率(train_acc
)和损失(train_loss
),累加,并计算平均的正确率和损失:
# 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
3. 编写测试函数
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
4. 正式训练
1.
model.train()
训练模式当调用
model.train()
时,模型中的一些层(如Dropout层、BatchNorm层等)会进行特定的操作以适应训练环境。例如,Dropout层会在训练过程中随机丢弃一部分神经元,以防止过拟合;BatchNorm层会根据当前批次的数据计算均值和方差,并进行归一化操作;2.
model.eval()
评估模式
model.eval()
的作用是不启用 Batch Normalization 和 Dropout。
epochs = 20 #训练20轮次
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')
四、结果可视化
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()