目录
- 前言
- 一、我的环境
- 二、代码实现
-
- [1. 前期准备](#1. 前期准备)
-
- [1.1 设置GPU](#1.1 设置GPU)
- [1.2 导入数据](#1.2 导入数据)
- [1.3 划分数据集](#1.3 划分数据集)
- [2. 构建简单的CNN网络](#2. 构建简单的CNN网络)
- [3. 训练模型](#3. 训练模型)
-
- [3.1 设置超参数](#3.1 设置超参数)
- [3.2 编写训练函数](#3.2 编写训练函数)
- [3.3 编写测试函数](#3.3 编写测试函数)
- [3.4 正式训练](#3.4 正式训练)
- [4. 结果可视化](#4. 结果可视化)
-
- [4.1 Loss与Accuracy图](#4.1 Loss与Accuracy图)
- [4.2 指定图片进行预测](#4.2 指定图片进行预测)
- [5. 保存并加载模型](#5. 保存并加载模型)
- 三、学习体会
前言
- 🍨 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
一、我的环境
●
🍺要求:
- 训练过程中保存效果最好的模型参数。
- 加载最佳模型参数识别本地的一张图片。
- 调整网络结构使测试集accuracy到达88%(重点)。
🍻拔高(可选):
- 调整模型参数并观察测试集的准确率变化。
- 尝试设置动态学习率。
- 测试集accuracy到达90%。
本周的代码相对于上周增加指定图片预测与保存并加载模型这个两个模块,在学习这个两知识点后,时间有余的同学请自由探索更佳的模型结构以提升模型是识别准确率,模型的搭建是深度学习程度的重点。
🏡 我的环境:
- 电脑系统:Windows 11
- 语言环境:Python 3.9.7
- 编辑器:Jupyter Lab
- 深度学习环境:TensorFlow2.4.1
二、代码实现
1. 前期准备
1.1 设置GPU
python
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")
device
1.2 导入数据
python
import os,PIL,random,pathlib
data_dir = './4-data/'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[1] for path in data_paths]
classeNames
● 第一步:使用pathlib.Path()函数将字符串类型的文件夹路径转换为pathlib.Path对象。
● 第二步:使用glob()方法获取data_dir路径下的所有文件路径,并以列表形式存储在data_paths中。
● 第三步:通过split()函数对data_paths中的每个文件路径执行分割操作,获得各个文件所属的类别名称,并存储在classeNames中
● 第四步:打印classeNames列表,显示每个文件所属的类别名称。
python
total_datadir = './4-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)
total_data
python
total_data.class_to_idx
total_data.class_to_idx是一个存储了数据集类别和对应索引的字典。在PyTorch的ImageFolder数据加载器中,根据数据集文件夹的组织结构,每个文件夹代表一个类别,class_to_idx字典将每个类别名称映射为一个数字索引。
具体来说,如果数据集文件夹包含两个子文件夹,比如Monkeypox和Others,class_to_idx字典将返回类似以下的映射关系:{'Monkeypox': 0, 'Others': 1}
1.3 划分数据集
python
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
python
train_size,test_size
python
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)
python
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
torch.utils.data.DataLoader()参数详解
torch.utils.data.DataLoader 是 PyTorch 中用于加载和管理数据的一个实用工具类。它允许你以小批次的方式迭代你的数据集,这对于训练神经网络和其他机器学习任务非常有用。DataLoader 构造函数接受多个参数,下面是一些常用的参数及其解释:
- 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。
2. 构建简单的CNN网络
网络结构图(可单击放大查看):
python
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
"""
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)))
x = F.relu(self.bn2(self.conv2(x)))
x = self.pool(x)
x = F.relu(self.bn4(self.conv4(x)))
x = F.relu(self.bn5(self.conv5(x)))
x = self.pool(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)
model
3. 训练模型
3.1 设置超参数
python
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)
3.2 编写训练函数
python
# 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
3.3 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
python
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
3.4 正式训练
python
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')
4. 结果可视化
4.1 Loss与Accuracy图
python
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()
4.2 指定图片进行预测
⭐torch.squeeze()详解
对数据的维度进行压缩,去掉维数为1的的维度
函数原型:
torch.squeeze(input, dim=None, *, out=None)
关键参数说明:
● input (Tensor):输入Tensor
● dim (int, optional):如果给定,输入将只在这个维度上被压缩
实战案例:
python
>>> x = torch.zeros(2, 1, 2, 1, 2)
>>> x.size()
torch.Size([2, 1, 2, 1, 2])
>>> y = torch.squeeze(x)
>>> y.size()
torch.Size([2, 2, 2])
>>> y = torch.squeeze(x, 0)
>>> y.size()
torch.Size([2, 1, 2, 1, 2])
>>> y = torch.squeeze(x, 1)
>>> y.size()
torch.Size([2, 2, 1, 2])
⭐torch.unsqueeze()
对数据维度进行扩充。给指定位置加上维数为一的维度
函数原型:
torch.unsqueeze(input, dim)
关键参数说明:
● input (Tensor):输入Tensor
● dim (int):插入单例维度的索引
实战案例:
python
>>> x = torch.tensor([1, 2, 3, 4])
>>> torch.unsqueeze(x, 0)
tensor([[ 1, 2, 3, 4]])
>>> torch.unsqueeze(x, 1)
tensor([[ 1],
[ 2],
[ 3],
[ 4]])
python
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}')
python
# 预测训练集中的某张照片
predict_one_image(image_path='./4-data/Monkeypox/M01_01_00.jpg',
model=model,
transform=train_transforms,
classes=classes)
5. 保存并加载模型
python
# 模型保存
PATH = './model.pth' # 保存的参数文件名
torch.save(model.state_dict(), PATH)
# 将参数加载到model当中
model.load_state_dict(torch.load(PATH, map_location=device))
三、学习体会
python
import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision.transforms as transforms
import torchvision
from torchvision import datasets
import pathlib
# 设置设备
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using {} device".format(device))
# 数据路径
data_dir = 'F:/boshiqijian/kechengxuexi/Deep Learning/365xunlianying/data/p4data/'
data_dir = pathlib.Path(data_dir)
total_datadir = 'F:/boshiqijian/kechengxuexi/Deep Learning/365xunlianying/data/p4data/'
# 类别名称
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[-1] for path in data_paths] # 改进路径分割
print("Class Names:", classeNames)
# 数据增强与预处理
train_transforms = transforms.Compose([
transforms.Resize([224, 224]),
transforms.RandomHorizontalFlip(), # 随机水平翻转
transforms.RandomRotation(15), # 随机旋转
transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1), # 颜色抖动
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
# 测试集的标准化变换
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])
])
# 加载数据集
total_data = datasets.ImageFolder(total_datadir, transform=train_transforms)
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.dataset.transform = train_transforms
test_dataset.dataset.transform = test_transforms
# 数据加载器
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=False, num_workers=1)
# 改进后的网络结构
class ImprovedNetwork(nn.Module):
def __init__(self):
super(ImprovedNetwork, self).__init__()
self.conv1 = nn.Conv2d(3, 32, kernel_size=3, stride=1, padding=1)
self.bn1 = nn.BatchNorm2d(32)
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1)
self.bn2 = nn.BatchNorm2d(64)
self.pool = nn.MaxPool2d(2, 2)
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1)
self.bn3 = nn.BatchNorm2d(128)
self.conv4 = nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1)
self.bn4 = nn.BatchNorm2d(128)
self.dropout = nn.Dropout(0.5)
self.fc1 = nn.Linear(128 * 56 * 56, 512)
self.fc2 = nn.Linear(512, len(classeNames))
def forward(self, x):
x = F.relu(self.bn1(self.conv1(x)))
x = F.relu(self.bn2(self.conv2(x)))
x = self.pool(x)
x = F.relu(self.bn3(self.conv3(x)))
x = F.relu(self.bn4(self.conv4(x)))
x = self.pool(x)
x = x.view(-1, 128 * 56 * 56)
x = F.relu(self.fc1(x))
x = self.dropout(x)
x = self.fc2(x)
return x
# 实例化模型
model = ImprovedNetwork().to(device)
print(model)
# 损失函数与优化器
loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-3
opt = torch.optim.Adam(model.parameters(), lr=learn_rate)
# 训练函数
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset)
num_batches = len(dataloader)
train_loss, train_acc = 0, 0
model.train()
for X, y in dataloader:
X, y = X.to(device), y.to(device)
# 前向传播
pred = model(X)
loss = loss_fn(pred, y)
# 反向传播
optimizer.zero_grad()
loss.backward()
optimizer.step()
# 记录loss和准确率
train_loss += loss.item()
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss /= num_batches
train_acc /= size
return train_acc, train_loss
# 测试函数
def test(dataloader, model, loss_fn):
size = len(dataloader.dataset)
num_batches = len(dataloader)
test_loss, test_acc = 0, 0
model.eval()
with torch.no_grad():
for X, y in dataloader:
X, y = X.to(device), y.to(device)
pred = model(X)
test_loss += loss_fn(pred, y).item()
test_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
test_loss /= num_batches
test_acc /= size
return test_acc, test_loss
# 训练模型
epochs = 20
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)
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)
print(f"Epoch {epoch+1}, Train_acc: {epoch_train_acc*100:.1f}%, Train_loss: {epoch_train_loss:.4f}, "
f"Test_acc: {epoch_test_acc*100:.1f}%, Test_loss: {epoch_test_loss:.4f}")
print("Training Complete!")