当前位置： 首页 > news >正文

深度学习每周学习总结J8（Inception V1 算法实战与解析 - 猴痘识别）

news 来源：原创 2025/6/4 18:01:46
  • 🍨 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

目录

      • 0. 总结
      • Inception V1 简介
      • 1. 设置GPU
      • 2. 导入数据及处理部分
      • 3. 划分数据集
      • 4. 模型构建部分
      • 5. 设置超参数：定义损失函数，学习率，以及根据学习率定义优化器等
      • 6. 训练函数
      • 7. 测试函数
      • 8. 正式训练
      • 9. 结果可视化
      • 10. 模型的保存
      • 11.使用训练好的模型进行预测

0. 总结

数据导入及处理部分：本次数据导入没有使用torchvision自带的数据集，需要将原始数据进行处理包括数据导入，查看数据分类情况，定义transforms，进行数据类型转换等操作。

划分数据集：划定训练集测试集后，再使用torch.utils.data中的DataLoader()分别加载上一步处理好的训练及测试数据，查看批处理维度.

模型构建部分：Inception V1

设置超参数：在这之前需要定义损失函数，学习率（动态学习率），以及根据学习率定义优化器（例如SGD随机梯度下降），用来在训练中更新参数，最小化损失函数。

定义训练函数：函数的传入的参数有四个，分别是设置好的DataLoader(),定义好的模型，损失函数，优化器。函数内部初始化损失准确率为0，接着开始循环，使用DataLoader()获取一个批次的数据，对这个批次的数据带入模型得到预测值，然后使用损失函数计算得到损失值。接下来就是进行反向传播以及使用优化器优化参数，梯度清零放在反向传播之前或者是使用优化器优化之后都是可以的，一般是默认放在反向传播之前。

定义测试函数：函数传入的参数相比训练函数少了优化器，只需传入设置好的DataLoader(),定义好的模型，损失函数。此外除了处理批次数据时无需再设置梯度清零、返向传播以及优化器优化参数，其余部分均和训练函数保持一致。

训练过程：定义训练次数，有几次就使用整个数据集进行几次训练，初始化四个空list分别存储每次训练及测试的准确率及损失。使用model.train()开启训练模式，调用训练函数得到准确率及损失。使用model.eval()将模型设置为评估模式，调用测试函数得到准确率及损失。接着就是将得到的训练及测试的准确率及损失存储到相应list中并合并打印出来，得到每一次整体训练后的准确率及损失。

结果可视化

模型的保存，调取及使用。在PyTorch中，通常使用 torch.save(model.state_dict(), ‘model.pth’) 保存模型的参数，使用 model.load_state_dict(torch.load(‘model.pth’)) 加载参数。

需要改进优化的地方：确保模型和数据的一致性，都存到GPU或者CPU;注意numclasses不要直接用默认的1000，需要根据实际数据集改进；实例化模型也要注意numclasses这个参数；此外注意测试模型需要用（3,224,224）3表示通道数，这和tensorflow定义的顺序是不用的（224,224,3），做代码转换时需要注意。

关于调优（十分重要）：本次将测试集准确率提升到了96.03%（随机种子设置为42）
1：使用多卡不一定比单卡效果好，需要继续调优
2：本次微调参数主要调整了两点一是初始学习率从1e-4 增大为了3e-4;其次是原来图片预处理只加入了随机水平翻转，本次加入了小角度的随机翻转，随机缩放剪裁，光照变化等，发现有更好的效果。测试集准确率有了很大的提升。从训练后的准确率图像也可以看到，训练准确率和测试准确率很接近甚至能够超过。之前没有做这个改进之前，都是训练准确率远大于测试准确率。

关键代码示例：

import torchvision.transforms as transforms# 定义猴痘识别的 transforms
train_transforms = transforms.Compose([transforms.Resize([224, 224]),            # 统一图片尺寸transforms.RandomHorizontalFlip(p=0.5),  # 随机水平翻转transforms.RandomRotation(degrees=15),   # 小角度随机旋转transforms.RandomResizedCrop(size=224, scale=(0.8, 1.2)),  # 随机缩放裁剪transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.1),  # 光照变化transforms.ToTensor(),                   # 转换为 Tensor 格式transforms.Normalize(                    # 标准化mean=[0.485, 0.456, 0.406],std=[0.229, 0.224, 0.225])
])

Inception V1 简介

什么是Inception V1？

Inception V1，也被称为GoogLeNet，是Google在2014年ILSVRC比赛中提出的一种卷积神经网络（CNN）架构，并且在比赛中获得了冠军。与当时流行的VGGNet相比，Inception V1在保持相似性能的同时，显著减少了参数数量，从而提高了计算效率。

Inception Module的核心思想

Inception V1的核心是Inception Module，它通过并行的卷积操作在同一层提取不同尺度的特征。这种设计不仅增加了网络的深度，还有效地捕捉了多种特征信息。

具体来说，一个Inception Module通常包含以下几个分支：

  1. 1x1卷积分支：用于降低输入特征图的通道数，减少计算量。
  2. 1x1卷积后接3x3卷积分支：先用1x1卷积降维，再进行3x3卷积提取特征。
  3. 1x1卷积后接5x5卷积分支：类似于3x3分支，但使用更大的卷积核以捕捉更大范围的特征。
  4. 3x3最大池化后接1x1卷积分支：先进行池化操作，再用1x1卷积进行特征整合。

通过将这些分支的输出在通道维度上拼接，Inception Module能够在同一层次上整合多种尺度的信息，提升模型的表达能力。

1x1卷积的作用

1x1卷积主要用于降维，即减少特征图的通道数。这不仅降低了网络的参数量和计算量，还间接增加了网络的深度，有助于提升模型性能。例如：

  • 原始输入：100x100x128
  • 经过1x1卷积降维到32通道，再进行5x5卷积，输出仍为100x100x256
  • 参数量由原来的约8.192×10⁹降低到2.048×10⁹

辅助分类器

Inception V1还引入了辅助分类器，主要有两个作用：

  1. 缓解梯度消失：通过在中间层添加分类器，帮助梯度更好地传播。
  2. 模型融合：将中间层的输出用于分类，增强模型的泛化能力。

不过，在实际应用中，这些辅助分类器通常在训练过程中使用，推理时会被去掉。

import torch
import torch.nn as nn
import torchvision
from torchvision import datasets,transforms
from torch.utils.data import DataLoader
import torchvision.models as models
import torch.nn.functional as F
from collections import OrderedDict import os,PIL,pathlib
import matplotlib.pyplot as plt
import warningswarnings.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

1. 设置GPU

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

device(type='cuda')

2. 导入数据及处理部分

# 获取数据分布情况
path_dir = './data/mpox_recognize/'
path_dir = pathlib.Path(path_dir)paths = list(path_dir.glob('*'))
# classNames = [str(path).split("\\")[-1] for path in paths] # ['Bananaquit', 'Black Skimmer', 'Black Throated Bushtiti', 'Cockatoo']
classNames = [path.parts[-1] for path in paths]
classNames

['Monkeypox', 'Others']

# 定义transforms 并处理数据
# 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 装换为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],    # 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。
#         std = [0.229,0.224,0.225]
#     )
# ])# 定义猴痘识别的 transforms 并处理数据
train_transforms = transforms.Compose([transforms.Resize([224, 224]),            # 统一图片尺寸transforms.RandomHorizontalFlip(p=0.5),  # 随机水平翻转transforms.RandomRotation(degrees=15),   # 小角度随机旋转transforms.RandomResizedCrop(size=224, scale=(0.8, 1.2)),  # 随机缩放裁剪transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.1),  # 光照变化transforms.ToTensor(),                   # 转换为 Tensor 格式transforms.Normalize(                    # 标准化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/mpox_recognize/',transform = train_transforms)
total_data

Dataset ImageFolderNumber of datapoints: 2142Root location: ./data/mpox_recognize/StandardTransform
Transform: Compose(Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=True)RandomHorizontalFlip(p=0.5)RandomRotation(degrees=[-15.0, 15.0], interpolation=nearest, expand=False, fill=0)RandomResizedCrop(size=(224, 224), scale=(0.8, 1.2), ratio=(0.75, 1.3333), interpolation=bilinear, antialias=True)ColorJitter(brightness=(0.8, 1.2), contrast=(0.8, 1.2), saturation=(0.9, 1.1), hue=None)ToTensor()Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]))

3. 划分数据集

# 设置随机种子
torch.manual_seed(42)# 划分数据集
train_size = int(len(total_data) * 0.8)
test_size = len(total_data) - train_sizetrain_dataset,test_dataset = torch.utils.data.random_split(total_data,[train_size,test_size])
train_dataset,test_dataset

(<torch.utils.data.dataset.Subset at 0x1b854727580>,<torch.utils.data.dataset.Subset at 0x1b854727c40>)

# 定义DataLoader用于数据集的加载batch_size = 32 # 如使用多显卡，请确保 batch_size 是显卡数量的倍数。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
)
# 观察数据维度
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
Shape of X [N,C,H,W]:  torch.Size([32, 3, 224, 224])
Shape of y:  torch.Size([32]) torch.int64

4. 模型构建部分

import torch
import torch.nn as nn
import torch.nn.functional as Fclass inception_block(nn.Module):def __init__(self, in_channels, ch1x1, ch3x3red, ch3x3, ch5x5red, ch5x5, pool_proj):super(inception_block, self).__init__()# 1x1 conv branchself.branch1 = nn.Sequential(nn.Conv2d(in_channels, ch1x1, kernel_size=1),nn.BatchNorm2d(ch1x1),nn.ReLU(inplace=True))# 1x1 conv -> 3x3 conv branchself.branch2 = nn.Sequential(nn.Conv2d(in_channels, ch3x3red, kernel_size=1),nn.BatchNorm2d(ch3x3red),nn.ReLU(inplace=True),nn.Conv2d(ch3x3red, ch3x3, kernel_size=3, padding=1),nn.BatchNorm2d(ch3x3),nn.ReLU(inplace=True))# 1x1 conv -> 5x5 conv branchself.branch3 = nn.Sequential(nn.Conv2d(in_channels, ch5x5red, kernel_size=1),nn.BatchNorm2d(ch5x5red),nn.ReLU(inplace=True),nn.Conv2d(ch5x5red, ch5x5, kernel_size=5, padding=2),nn.BatchNorm2d(ch5x5),nn.ReLU(inplace=True))# 3x3 max pooling -> 1x1 conv branchself.branch4 = nn.Sequential(nn.MaxPool2d(kernel_size=3, stride=1, padding=1),nn.Conv2d(in_channels, pool_proj, kernel_size=1),nn.BatchNorm2d(pool_proj),nn.ReLU(inplace=True))def forward(self, x):# Compute forward pass through all branches and concatenate the output feature mapsbranch1_output = self.branch1(x)branch2_output = self.branch2(x)branch3_output = self.branch3(x)branch4_output = self.branch4(x)outputs = [branch1_output, branch2_output, branch3_output, branch4_output]return torch.cat(outputs, 1)class InceptionV1(nn.Module):def __init__(self, num_classes=1000):super(InceptionV1, self).__init__()self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3)self.maxpool1 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.conv2 = nn.Conv2d(64, 64, kernel_size=1, stride=1, padding=0)self.conv3 = nn.Conv2d(64, 192, kernel_size=3, stride=1, padding=1)self.maxpool2 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.inception3a = inception_block(192, 64, 96, 128, 16, 32, 32)self.inception3b = inception_block(256, 128, 128, 192, 32, 96, 64)self.maxpool3    = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.inception4a = inception_block(480, 192, 96, 208, 16, 48, 64)self.inception4b = inception_block(512, 160, 112, 224, 24, 64, 64)self.inception4c = inception_block(512, 128, 128, 256, 24, 64, 64)self.inception4d = inception_block(512, 112, 144, 288, 32, 64, 64)self.inception4e = inception_block(528, 256, 160, 320, 32, 128, 128)self.maxpool4    = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)self.inception5a = inception_block(832, 256, 160, 320, 32, 128, 128)self.inception5b=nn.Sequential(inception_block(832, 384, 192, 384, 48, 128, 128),nn.AvgPool2d(kernel_size=7,stride=1,padding=0),nn.Dropout(0.4))# 全连接层前的池化层： 在Inception V1中，最后一个Inception模块后通常会有一个全局平均池化层，# 以减少特征维度。你可以在inception5b后添加：self.avgpool = nn.AdaptiveAvgPool2d((1, 1))# 全连接网络层，用于分类self.classifier = nn.Sequential(nn.Linear(in_features=1024, out_features=1024),nn.ReLU(),nn.Linear(in_features=1024, out_features=num_classes),nn.Softmax(dim=1))def forward(self, x):x = self.conv1(x)x = F.relu(x)x = self.maxpool1(x)x = self.conv2(x)x = F.relu(x)x = self.conv3(x)x = F.relu(x)x = self.maxpool2(x)x = self.inception3a(x)x = self.inception3b(x)x = self.maxpool3(x)x = self.inception4a(x)x = self.inception4b(x)x = self.inception4c(x)x = self.inception4d(x)x = self.inception4e(x)x = self.maxpool4(x)x = self.inception5a(x)x = self.inception5b(x)x = self.avgpool(x) # 全连接层前的池化层x = torch.flatten(x, start_dim=1)x = self.classifier(x)return x
model = InceptionV1(num_classes=len(classNames)).to(device)
model

InceptionV1((conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3))(maxpool1): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(conv2): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))(conv3): Conv2d(64, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(maxpool2): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(inception3a): inception_block((branch1): Sequential((0): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(96, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(192, 16, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(16, 32, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(192, 32, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception3b): inception_block((branch1): Sequential((0): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(128, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(256, 32, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(32, 96, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(maxpool3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(inception4a): inception_block((branch1): Sequential((0): Conv2d(480, 192, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(480, 96, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(96, 208, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(208, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(480, 16, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(16, 48, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(480, 64, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception4b): inception_block((branch1): Sequential((0): Conv2d(512, 160, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(112, 224, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(224, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(24, 64, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception4c): inception_block((branch1): Sequential((0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(24, 64, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception4d): inception_block((branch1): Sequential((0): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(512, 144, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(144, 288, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(288, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(32, 64, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception4e): inception_block((branch1): Sequential((0): Conv2d(528, 256, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(528, 160, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(528, 32, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(32, 128, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(528, 128, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(maxpool4): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(inception5a): inception_block((branch1): Sequential((0): Conv2d(832, 256, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(832, 160, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(832, 32, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(32, 128, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(inception5b): Sequential((0): inception_block((branch1): Sequential((0): Conv2d(832, 384, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True))(branch2): Sequential((0): Conv2d(832, 192, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(4): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch3): Sequential((0): Conv2d(832, 48, kernel_size=(1, 1), stride=(1, 1))(1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(2): ReLU(inplace=True)(3): Conv2d(48, 128, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(5): ReLU(inplace=True))(branch4): Sequential((0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=False)(1): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1))(2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(3): ReLU(inplace=True)))(1): AvgPool2d(kernel_size=7, stride=1, padding=0)(2): Dropout(p=0.4, inplace=False))(avgpool): AdaptiveAvgPool2d(output_size=(1, 1))(classifier): Sequential((0): Linear(in_features=1024, out_features=1024, bias=True)(1): ReLU()(2): Linear(in_features=1024, out_features=2, bias=True)(3): Softmax(dim=1))
)

# 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, 112, 112]           9,472MaxPool2d-2           [-1, 64, 56, 56]               0Conv2d-3           [-1, 64, 56, 56]           4,160Conv2d-4          [-1, 192, 56, 56]         110,784MaxPool2d-5          [-1, 192, 28, 28]               0Conv2d-6           [-1, 64, 28, 28]          12,352BatchNorm2d-7           [-1, 64, 28, 28]             128ReLU-8           [-1, 64, 28, 28]               0Conv2d-9           [-1, 96, 28, 28]          18,528BatchNorm2d-10           [-1, 96, 28, 28]             192ReLU-11           [-1, 96, 28, 28]               0Conv2d-12          [-1, 128, 28, 28]         110,720BatchNorm2d-13          [-1, 128, 28, 28]             256ReLU-14          [-1, 128, 28, 28]               0Conv2d-15           [-1, 16, 28, 28]           3,088BatchNorm2d-16           [-1, 16, 28, 28]              32ReLU-17           [-1, 16, 28, 28]               0Conv2d-18           [-1, 32, 28, 28]          12,832BatchNorm2d-19           [-1, 32, 28, 28]              64ReLU-20           [-1, 32, 28, 28]               0MaxPool2d-21          [-1, 192, 28, 28]               0Conv2d-22           [-1, 32, 28, 28]           6,176BatchNorm2d-23           [-1, 32, 28, 28]              64ReLU-24           [-1, 32, 28, 28]               0inception_block-25          [-1, 256, 28, 28]               0Conv2d-26          [-1, 128, 28, 28]          32,896BatchNorm2d-27          [-1, 128, 28, 28]             256ReLU-28          [-1, 128, 28, 28]               0Conv2d-29          [-1, 128, 28, 28]          32,896BatchNorm2d-30          [-1, 128, 28, 28]             256ReLU-31          [-1, 128, 28, 28]               0Conv2d-32          [-1, 192, 28, 28]         221,376BatchNorm2d-33          [-1, 192, 28, 28]             384ReLU-34          [-1, 192, 28, 28]               0Conv2d-35           [-1, 32, 28, 28]           8,224BatchNorm2d-36           [-1, 32, 28, 28]              64ReLU-37           [-1, 32, 28, 28]               0Conv2d-38           [-1, 96, 28, 28]          76,896BatchNorm2d-39           [-1, 96, 28, 28]             192ReLU-40           [-1, 96, 28, 28]               0MaxPool2d-41          [-1, 256, 28, 28]               0Conv2d-42           [-1, 64, 28, 28]          16,448BatchNorm2d-43           [-1, 64, 28, 28]             128ReLU-44           [-1, 64, 28, 28]               0inception_block-45          [-1, 480, 28, 28]               0MaxPool2d-46          [-1, 480, 14, 14]               0Conv2d-47          [-1, 192, 14, 14]          92,352BatchNorm2d-48          [-1, 192, 14, 14]             384ReLU-49          [-1, 192, 14, 14]               0Conv2d-50           [-1, 96, 14, 14]          46,176BatchNorm2d-51           [-1, 96, 14, 14]             192ReLU-52           [-1, 96, 14, 14]               0Conv2d-53          [-1, 208, 14, 14]         179,920BatchNorm2d-54          [-1, 208, 14, 14]             416ReLU-55          [-1, 208, 14, 14]               0Conv2d-56           [-1, 16, 14, 14]           7,696BatchNorm2d-57           [-1, 16, 14, 14]              32ReLU-58           [-1, 16, 14, 14]               0Conv2d-59           [-1, 48, 14, 14]          19,248BatchNorm2d-60           [-1, 48, 14, 14]              96ReLU-61           [-1, 48, 14, 14]               0MaxPool2d-62          [-1, 480, 14, 14]               0Conv2d-63           [-1, 64, 14, 14]          30,784BatchNorm2d-64           [-1, 64, 14, 14]             128ReLU-65           [-1, 64, 14, 14]               0inception_block-66          [-1, 512, 14, 14]               0Conv2d-67          [-1, 160, 14, 14]          82,080BatchNorm2d-68          [-1, 160, 14, 14]             320ReLU-69          [-1, 160, 14, 14]               0Conv2d-70          [-1, 112, 14, 14]          57,456BatchNorm2d-71          [-1, 112, 14, 14]             224ReLU-72          [-1, 112, 14, 14]               0Conv2d-73          [-1, 224, 14, 14]         226,016BatchNorm2d-74          [-1, 224, 14, 14]             448ReLU-75          [-1, 224, 14, 14]               0Conv2d-76           [-1, 24, 14, 14]          12,312BatchNorm2d-77           [-1, 24, 14, 14]              48ReLU-78           [-1, 24, 14, 14]               0Conv2d-79           [-1, 64, 14, 14]          38,464BatchNorm2d-80           [-1, 64, 14, 14]             128ReLU-81           [-1, 64, 14, 14]               0MaxPool2d-82          [-1, 512, 14, 14]               0Conv2d-83           [-1, 64, 14, 14]          32,832BatchNorm2d-84           [-1, 64, 14, 14]             128ReLU-85           [-1, 64, 14, 14]               0inception_block-86          [-1, 512, 14, 14]               0Conv2d-87          [-1, 128, 14, 14]          65,664BatchNorm2d-88          [-1, 128, 14, 14]             256ReLU-89          [-1, 128, 14, 14]               0Conv2d-90          [-1, 128, 14, 14]          65,664BatchNorm2d-91          [-1, 128, 14, 14]             256ReLU-92          [-1, 128, 14, 14]               0Conv2d-93          [-1, 256, 14, 14]         295,168BatchNorm2d-94          [-1, 256, 14, 14]             512ReLU-95          [-1, 256, 14, 14]               0Conv2d-96           [-1, 24, 14, 14]          12,312BatchNorm2d-97           [-1, 24, 14, 14]              48ReLU-98           [-1, 24, 14, 14]               0Conv2d-99           [-1, 64, 14, 14]          38,464BatchNorm2d-100           [-1, 64, 14, 14]             128ReLU-101           [-1, 64, 14, 14]               0MaxPool2d-102          [-1, 512, 14, 14]               0Conv2d-103           [-1, 64, 14, 14]          32,832BatchNorm2d-104           [-1, 64, 14, 14]             128ReLU-105           [-1, 64, 14, 14]               0inception_block-106          [-1, 512, 14, 14]               0Conv2d-107          [-1, 112, 14, 14]          57,456BatchNorm2d-108          [-1, 112, 14, 14]             224ReLU-109          [-1, 112, 14, 14]               0Conv2d-110          [-1, 144, 14, 14]          73,872BatchNorm2d-111          [-1, 144, 14, 14]             288ReLU-112          [-1, 144, 14, 14]               0Conv2d-113          [-1, 288, 14, 14]         373,536BatchNorm2d-114          [-1, 288, 14, 14]             576ReLU-115          [-1, 288, 14, 14]               0Conv2d-116           [-1, 32, 14, 14]          16,416BatchNorm2d-117           [-1, 32, 14, 14]              64ReLU-118           [-1, 32, 14, 14]               0Conv2d-119           [-1, 64, 14, 14]          51,264BatchNorm2d-120           [-1, 64, 14, 14]             128ReLU-121           [-1, 64, 14, 14]               0MaxPool2d-122          [-1, 512, 14, 14]               0Conv2d-123           [-1, 64, 14, 14]          32,832BatchNorm2d-124           [-1, 64, 14, 14]             128ReLU-125           [-1, 64, 14, 14]               0inception_block-126          [-1, 528, 14, 14]               0Conv2d-127          [-1, 256, 14, 14]         135,424BatchNorm2d-128          [-1, 256, 14, 14]             512ReLU-129          [-1, 256, 14, 14]               0Conv2d-130          [-1, 160, 14, 14]          84,640BatchNorm2d-131          [-1, 160, 14, 14]             320ReLU-132          [-1, 160, 14, 14]               0Conv2d-133          [-1, 320, 14, 14]         461,120BatchNorm2d-134          [-1, 320, 14, 14]             640ReLU-135          [-1, 320, 14, 14]               0Conv2d-136           [-1, 32, 14, 14]          16,928BatchNorm2d-137           [-1, 32, 14, 14]              64ReLU-138           [-1, 32, 14, 14]               0Conv2d-139          [-1, 128, 14, 14]         102,528BatchNorm2d-140          [-1, 128, 14, 14]             256ReLU-141          [-1, 128, 14, 14]               0MaxPool2d-142          [-1, 528, 14, 14]               0Conv2d-143          [-1, 128, 14, 14]          67,712BatchNorm2d-144          [-1, 128, 14, 14]             256ReLU-145          [-1, 128, 14, 14]               0inception_block-146          [-1, 832, 14, 14]               0MaxPool2d-147            [-1, 832, 7, 7]               0Conv2d-148            [-1, 256, 7, 7]         213,248BatchNorm2d-149            [-1, 256, 7, 7]             512ReLU-150            [-1, 256, 7, 7]               0Conv2d-151            [-1, 160, 7, 7]         133,280BatchNorm2d-152            [-1, 160, 7, 7]             320ReLU-153            [-1, 160, 7, 7]               0Conv2d-154            [-1, 320, 7, 7]         461,120BatchNorm2d-155            [-1, 320, 7, 7]             640ReLU-156            [-1, 320, 7, 7]               0Conv2d-157             [-1, 32, 7, 7]          26,656BatchNorm2d-158             [-1, 32, 7, 7]              64ReLU-159             [-1, 32, 7, 7]               0Conv2d-160            [-1, 128, 7, 7]         102,528BatchNorm2d-161            [-1, 128, 7, 7]             256ReLU-162            [-1, 128, 7, 7]               0MaxPool2d-163            [-1, 832, 7, 7]               0Conv2d-164            [-1, 128, 7, 7]         106,624BatchNorm2d-165            [-1, 128, 7, 7]             256ReLU-166            [-1, 128, 7, 7]               0inception_block-167            [-1, 832, 7, 7]               0Conv2d-168            [-1, 384, 7, 7]         319,872BatchNorm2d-169            [-1, 384, 7, 7]             768ReLU-170            [-1, 384, 7, 7]               0Conv2d-171            [-1, 192, 7, 7]         159,936BatchNorm2d-172            [-1, 192, 7, 7]             384ReLU-173            [-1, 192, 7, 7]               0Conv2d-174            [-1, 384, 7, 7]         663,936BatchNorm2d-175            [-1, 384, 7, 7]             768ReLU-176            [-1, 384, 7, 7]               0Conv2d-177             [-1, 48, 7, 7]          39,984BatchNorm2d-178             [-1, 48, 7, 7]              96ReLU-179             [-1, 48, 7, 7]               0Conv2d-180            [-1, 128, 7, 7]         153,728BatchNorm2d-181            [-1, 128, 7, 7]             256ReLU-182            [-1, 128, 7, 7]               0MaxPool2d-183            [-1, 832, 7, 7]               0Conv2d-184            [-1, 128, 7, 7]         106,624BatchNorm2d-185            [-1, 128, 7, 7]             256ReLU-186            [-1, 128, 7, 7]               0inception_block-187           [-1, 1024, 7, 7]               0AvgPool2d-188           [-1, 1024, 1, 1]               0Dropout-189           [-1, 1024, 1, 1]               0
AdaptiveAvgPool2d-190           [-1, 1024, 1, 1]               0Linear-191                 [-1, 1024]       1,049,600ReLU-192                 [-1, 1024]               0Linear-193                    [-1, 2]           2,050Softmax-194                    [-1, 2]               0
================================================================
Total params: 7,039,122
Trainable params: 7,039,122
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 69.62
Params size (MB): 26.85
Estimated Total Size (MB): 97.05
----------------------------------------------------------------

5. 设置超参数：定义损失函数，学习率，以及根据学习率定义优化器等

# loss_fn = nn.CrossEntropyLoss() # 创建损失函数# learn_rate = 1e-3 # 初始学习率
# def adjust_learning_rate(optimizer,epoch,start_lr):
#     # 每两个epoch 衰减到原来的0.98
#     lr = start_lr * (0.92 ** (epoch//2))
#     for param_group in optimizer.param_groups:
#         param_group['lr'] = lr# optimizer = torch.optim.Adam(model.parameters(),lr=learn_rate)
# 调用官方接口示例
loss_fn = nn.CrossEntropyLoss()# learn_rate = 1e-4  
learn_rate = 3e-4
lambda1 = lambda epoch:(0.92**(epoch//2))optimizer = torch.optim.Adam(model.parameters(),lr = learn_rate)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer,lr_lambda=lambda1) # 选定调整方法

6. 训练函数

# 训练函数
def train(dataloader,model,loss_fn,optimizer):size = len(dataloader.dataset) # 训练集大小num_batches = len(dataloader) # 批次数目train_loss,train_acc = 0,0for 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losstrain_acc += (pred.argmax(1)==y).type(torch.float).sum().item()train_loss += loss.item()train_acc /= sizetrain_loss /= num_batchesreturn train_acc,train_loss

7. 测试函数

# 测试函数
def test(dataloader,model,loss_fn):size = len(dataloader.dataset)num_batches = len(dataloader)test_acc,test_loss = 0,0with torch.no_grad():for X,y in dataloader:X,y = X.to(device),y.to(device)# 计算losspred = model(X)loss = loss_fn(pred,y)test_acc += (pred.argmax(1)==y).type(torch.float).sum().item()test_loss += loss.item()test_acc /= sizetest_loss /= num_batchesreturn test_acc,test_loss

8. 正式训练

import copyepochs = 60train_acc = []
train_loss = []
test_acc = []
test_loss = []best_acc = 0.0# 检查 GPU 可用性并打印设备信息
if torch.cuda.is_available():for i in range(torch.cuda.device_count()):print(f"GPU {i}: {torch.cuda.get_device_name(i)}")print(f"Initial Memory Allocated: {torch.cuda.memory_allocated(i)/1024**2:.2f} MB")print(f"Initial Memory Cached: {torch.cuda.memory_reserved(i)/1024**2:.2f} MB")
else:print("No GPU available. Using CPU.")# 多显卡设置 当前使用的是使用 PyTorch 自带的 DataParallel，后续如有需要可以设置为DistributedDataParallel，这是更加高效的方式
# 且多卡不一定比单卡效果就好，需要调整优化
# if torch.cuda.device_count() > 1:
#     print(f"Using {torch.cuda.device_count()} GPUs")
#     model = nn.DataParallel(model)
# model = model.to('cuda')for epoch in range(epochs):# 更新学习率——使用自定义学习率时使用# adjust_learning_rate(optimizer,epoch,learn_rate)model.train()epoch_train_acc,epoch_train_loss = train(train_dl,model,loss_fn,optimizer)scheduler.step() # 更新学习率——调用官方动态学习率时使用model.eval()epoch_test_acc,epoch_test_loss = test(test_dl,model,loss_fn)# 保存最佳模型到 best_modelif epoch_test_acc > best_acc:best_acc = epoch_test_accbest_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))# 实时监控 GPU 状态if torch.cuda.is_available():for i in range(torch.cuda.device_count()):print(f"GPU {i} Usage:")print(f"  Memory Allocated: {torch.cuda.memory_allocated(i)/1024**2:.2f} MB")print(f"  Memory Cached: {torch.cuda.memory_reserved(i)/1024**2:.2f} MB")print(f"  Max Memory Allocated: {torch.cuda.max_memory_allocated(i)/1024**2:.2f} MB")print(f"  Max Memory Cached: {torch.cuda.max_memory_reserved(i)/1024**2:.2f} MB")print('Done','best_acc: ',best_acc)

GPU 0: NVIDIA GeForce RTX 4070 Laptop GPU
Initial Memory Allocated: 335.65 MB
Initial Memory Cached: 586.00 MB
Epoch: 1,Train_acc:63.3%,Train_loss:0.645,Test_acc:64.8%,Test_loss:0.634,Lr:3.00E-04
GPU 0 Usage:Memory Allocated: 455.01 MBMemory Cached: 2072.00 MBMax Memory Allocated: 1845.06 MBMax Memory Cached: 2072.00 MB
Epoch: 2,Train_acc:63.6%,Train_loss:0.638,Test_acc:62.5%,Test_loss:0.670,Lr:2.76E-04
GPU 0 Usage:Memory Allocated: 454.59 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1872.34 MBMax Memory Cached: 2086.00 MB
Epoch: 3,Train_acc:67.1%,Train_loss:0.622,Test_acc:62.9%,Test_loss:0.651,Lr:2.76E-04
GPU 0 Usage:Memory Allocated: 454.37 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1874.41 MBMax Memory Cached: 2086.00 MB
Epoch: 4,Train_acc:66.1%,Train_loss:0.627,Test_acc:67.1%,Test_loss:0.621,Lr:2.54E-04
GPU 0 Usage:Memory Allocated: 453.73 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1874.41 MBMax Memory Cached: 2086.00 MB
Epoch: 5,Train_acc:68.6%,Train_loss:0.616,Test_acc:60.8%,Test_loss:0.683,Lr:2.54E-04
GPU 0 Usage:Memory Allocated: 453.00 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1874.41 MBMax Memory Cached: 2086.00 MB
Epoch: 6,Train_acc:68.5%,Train_loss:0.601,Test_acc:69.5%,Test_loss:0.602,Lr:2.34E-04
GPU 0 Usage:Memory Allocated: 454.46 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1874.41 MBMax Memory Cached: 2086.00 MB
Epoch: 7,Train_acc:72.2%,Train_loss:0.583,Test_acc:70.4%,Test_loss:0.601,Lr:2.34E-04
GPU 0 Usage:Memory Allocated: 454.09 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1874.41 MBMax Memory Cached: 2086.00 MB
Epoch: 8,Train_acc:72.6%,Train_loss:0.572,Test_acc:69.9%,Test_loss:0.607,Lr:2.15E-04
GPU 0 Usage:Memory Allocated: 453.72 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch: 9,Train_acc:75.8%,Train_loss:0.545,Test_acc:73.7%,Test_loss:0.567,Lr:2.15E-04
GPU 0 Usage:Memory Allocated: 454.52 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:10,Train_acc:75.8%,Train_loss:0.544,Test_acc:72.0%,Test_loss:0.584,Lr:1.98E-04
GPU 0 Usage:Memory Allocated: 454.52 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:11,Train_acc:76.6%,Train_loss:0.539,Test_acc:75.3%,Test_loss:0.542,Lr:1.98E-04
GPU 0 Usage:Memory Allocated: 455.12 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:12,Train_acc:78.6%,Train_loss:0.517,Test_acc:72.7%,Test_loss:0.574,Lr:1.82E-04
GPU 0 Usage:Memory Allocated: 455.12 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:13,Train_acc:78.2%,Train_loss:0.521,Test_acc:74.1%,Test_loss:0.569,Lr:1.82E-04
GPU 0 Usage:Memory Allocated: 455.12 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:14,Train_acc:78.1%,Train_loss:0.525,Test_acc:79.3%,Test_loss:0.509,Lr:1.67E-04
GPU 0 Usage:Memory Allocated: 454.52 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:15,Train_acc:83.0%,Train_loss:0.483,Test_acc:72.7%,Test_loss:0.575,Lr:1.67E-04
GPU 0 Usage:Memory Allocated: 454.52 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.13 MBMax Memory Cached: 2086.00 MB
Epoch:16,Train_acc:82.6%,Train_loss:0.482,Test_acc:75.3%,Test_loss:0.545,Lr:1.54E-04
GPU 0 Usage:Memory Allocated: 455.53 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.22 MBMax Memory Cached: 2086.00 MB
Epoch:17,Train_acc:83.1%,Train_loss:0.476,Test_acc:79.5%,Test_loss:0.506,Lr:1.54E-04
GPU 0 Usage:Memory Allocated: 454.47 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.22 MBMax Memory Cached: 2086.00 MB
Epoch:18,Train_acc:84.8%,Train_loss:0.457,Test_acc:83.4%,Test_loss:0.471,Lr:1.42E-04
GPU 0 Usage:Memory Allocated: 454.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.22 MBMax Memory Cached: 2086.00 MB
Epoch:19,Train_acc:84.5%,Train_loss:0.467,Test_acc:81.8%,Test_loss:0.495,Lr:1.42E-04
GPU 0 Usage:Memory Allocated: 455.60 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:20,Train_acc:85.2%,Train_loss:0.457,Test_acc:83.2%,Test_loss:0.467,Lr:1.30E-04
GPU 0 Usage:Memory Allocated: 455.02 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:21,Train_acc:86.0%,Train_loss:0.445,Test_acc:79.7%,Test_loss:0.503,Lr:1.30E-04
GPU 0 Usage:Memory Allocated: 455.60 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:22,Train_acc:86.2%,Train_loss:0.444,Test_acc:86.0%,Test_loss:0.454,Lr:1.20E-04
GPU 0 Usage:Memory Allocated: 453.39 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:23,Train_acc:87.0%,Train_loss:0.437,Test_acc:85.5%,Test_loss:0.452,Lr:1.20E-04
GPU 0 Usage:Memory Allocated: 453.02 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:24,Train_acc:87.9%,Train_loss:0.432,Test_acc:88.8%,Test_loss:0.427,Lr:1.10E-04
GPU 0 Usage:Memory Allocated: 454.34 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:25,Train_acc:88.5%,Train_loss:0.423,Test_acc:86.2%,Test_loss:0.435,Lr:1.10E-04
GPU 0 Usage:Memory Allocated: 454.33 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:26,Train_acc:89.3%,Train_loss:0.421,Test_acc:86.7%,Test_loss:0.436,Lr:1.01E-04
GPU 0 Usage:Memory Allocated: 454.33 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:27,Train_acc:90.0%,Train_loss:0.411,Test_acc:87.2%,Test_loss:0.435,Lr:1.01E-04
GPU 0 Usage:Memory Allocated: 454.33 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:28,Train_acc:90.4%,Train_loss:0.404,Test_acc:89.3%,Test_loss:0.424,Lr:9.34E-05
GPU 0 Usage:Memory Allocated: 453.29 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:29,Train_acc:90.3%,Train_loss:0.405,Test_acc:89.7%,Test_loss:0.411,Lr:9.34E-05
GPU 0 Usage:Memory Allocated: 455.36 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:30,Train_acc:89.6%,Train_loss:0.411,Test_acc:89.0%,Test_loss:0.424,Lr:8.59E-05
GPU 0 Usage:Memory Allocated: 455.31 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:31,Train_acc:91.9%,Train_loss:0.392,Test_acc:90.7%,Test_loss:0.412,Lr:8.59E-05
GPU 0 Usage:Memory Allocated: 453.24 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:32,Train_acc:91.8%,Train_loss:0.392,Test_acc:89.5%,Test_loss:0.420,Lr:7.90E-05
GPU 0 Usage:Memory Allocated: 453.27 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:33,Train_acc:91.7%,Train_loss:0.392,Test_acc:91.8%,Test_loss:0.387,Lr:7.90E-05
GPU 0 Usage:Memory Allocated: 455.28 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:34,Train_acc:91.0%,Train_loss:0.401,Test_acc:89.7%,Test_loss:0.410,Lr:7.27E-05
GPU 0 Usage:Memory Allocated: 454.91 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:35,Train_acc:91.6%,Train_loss:0.392,Test_acc:92.5%,Test_loss:0.389,Lr:7.27E-05
GPU 0 Usage:Memory Allocated: 453.79 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:36,Train_acc:92.8%,Train_loss:0.386,Test_acc:92.1%,Test_loss:0.387,Lr:6.69E-05
GPU 0 Usage:Memory Allocated: 453.79 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:37,Train_acc:91.9%,Train_loss:0.392,Test_acc:88.8%,Test_loss:0.422,Lr:6.69E-05
GPU 0 Usage:Memory Allocated: 453.79 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:38,Train_acc:93.2%,Train_loss:0.382,Test_acc:90.9%,Test_loss:0.405,Lr:6.15E-05
GPU 0 Usage:Memory Allocated: 453.79 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:39,Train_acc:93.1%,Train_loss:0.382,Test_acc:93.0%,Test_loss:0.380,Lr:6.15E-05
GPU 0 Usage:Memory Allocated: 455.30 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:40,Train_acc:93.1%,Train_loss:0.381,Test_acc:92.8%,Test_loss:0.386,Lr:5.66E-05
GPU 0 Usage:Memory Allocated: 455.30 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:41,Train_acc:93.5%,Train_loss:0.377,Test_acc:92.8%,Test_loss:0.387,Lr:5.66E-05
GPU 0 Usage:Memory Allocated: 455.30 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:42,Train_acc:93.8%,Train_loss:0.373,Test_acc:93.9%,Test_loss:0.374,Lr:5.21E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:43,Train_acc:94.3%,Train_loss:0.370,Test_acc:92.8%,Test_loss:0.381,Lr:5.21E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:44,Train_acc:93.8%,Train_loss:0.373,Test_acc:92.3%,Test_loss:0.394,Lr:4.79E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:45,Train_acc:94.5%,Train_loss:0.368,Test_acc:93.9%,Test_loss:0.367,Lr:4.79E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:46,Train_acc:94.3%,Train_loss:0.370,Test_acc:93.0%,Test_loss:0.385,Lr:4.41E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:47,Train_acc:94.6%,Train_loss:0.364,Test_acc:93.0%,Test_loss:0.378,Lr:4.41E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:48,Train_acc:93.2%,Train_loss:0.380,Test_acc:93.0%,Test_loss:0.383,Lr:4.06E-05
GPU 0 Usage:Memory Allocated: 453.05 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:49,Train_acc:93.9%,Train_loss:0.371,Test_acc:95.1%,Test_loss:0.361,Lr:4.06E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:50,Train_acc:94.6%,Train_loss:0.368,Test_acc:94.6%,Test_loss:0.367,Lr:3.73E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:51,Train_acc:94.3%,Train_loss:0.368,Test_acc:94.6%,Test_loss:0.365,Lr:3.73E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:52,Train_acc:94.9%,Train_loss:0.363,Test_acc:93.2%,Test_loss:0.376,Lr:3.43E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:53,Train_acc:95.2%,Train_loss:0.362,Test_acc:94.4%,Test_loss:0.362,Lr:3.43E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:54,Train_acc:96.4%,Train_loss:0.348,Test_acc:94.4%,Test_loss:0.373,Lr:3.16E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:55,Train_acc:96.5%,Train_loss:0.347,Test_acc:93.7%,Test_loss:0.371,Lr:3.16E-05
GPU 0 Usage:Memory Allocated: 454.08 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:56,Train_acc:95.6%,Train_loss:0.355,Test_acc:95.8%,Test_loss:0.355,Lr:2.91E-05
GPU 0 Usage:Memory Allocated: 453.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:57,Train_acc:95.4%,Train_loss:0.358,Test_acc:94.6%,Test_loss:0.363,Lr:2.91E-05
GPU 0 Usage:Memory Allocated: 453.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:58,Train_acc:95.6%,Train_loss:0.355,Test_acc:94.4%,Test_loss:0.369,Lr:2.67E-05
GPU 0 Usage:Memory Allocated: 453.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:59,Train_acc:96.5%,Train_loss:0.348,Test_acc:93.9%,Test_loss:0.372,Lr:2.67E-05
GPU 0 Usage:Memory Allocated: 453.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Epoch:60,Train_acc:96.8%,Train_loss:0.344,Test_acc:96.0%,Test_loss:0.355,Lr:2.46E-05
GPU 0 Usage:Memory Allocated: 453.93 MBMemory Cached: 2086.00 MBMax Memory Allocated: 1875.26 MBMax Memory Cached: 2086.00 MB
Done best_acc:  0.9603729603729604

9. 结果可视化

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 = 'Test Accuracy')
plt.plot(epochs_range,test_loss,label = 'Test Loss')
plt.legend(loc = 'lower right')
plt.title('Training and validation Loss')
plt.show()

请添加图片描述

​

10. 模型的保存

# 自定义模型保存
# 状态字典保存
torch.save(model.state_dict(),'./模型参数/J8_InceptionV1_model_state_dict.pth') # 仅保存状态字典# 定义模型用来加载参数
best_model = InceptionV1(num_classes=len(classNames)).to(device)best_model.load_state_dict(torch.load('./模型参数/J8_InceptionV1_model_state_dict.pth')) # 加载状态字典到模型
<All keys matched successfully>

11.使用训练好的模型进行预测

# 指定路径图片预测
from PIL import Image
import torchvision.transforms as transformsclasses = list(total_data.class_to_idx) # 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) # 展示待预测的图片test_img = transform(test_img)img = test_img.to(device).unsqueeze(0)model.eval()output = model(img)print(output) # 观察模型预测结果的输出数据_,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/mpox_recognize/Monkeypox/M01_01_04.jpg',model = model,transform = test_transforms,classes = classes)
tensor([[0.0015, 0.9985]], device='cuda:0', grad_fn=<SoftmaxBackward0>)
预测结果是:Others

classes

['Monkeypox', 'Others']

相关文章：

深度学习每周学习总结J8（Inception V1 算法实战与解析 - 猴痘识别）

&#x1f368; 本文为&#x1f517;365天深度学习训练营 中的学习记录博客&#x1f356; 原作者&#xff1a;K同学啊 | 接辅导、项目定制 目录 0. 总结Inception V1 简介1. 设置GPU2. 导入数据及处理部分3. 划分数据集4. 模型构建部分5. 设置超参数&#xff1a;定义损失函数&am…...

编程日记 2025/6/4 18:01:46

Django模板系统

1.常用语法 Django模板中只需要记两种特殊符号&#xff1a; {{ }}和 {% %} {{ }}表示变量&#xff0c;在模板渲染的时候替换成值&#xff0c;{% %}表示逻辑相关的操作。 2.变量 {{ 变量名 }} 变量名由字母数字和下划线组成。 点&#xff08;.&#xff09;在模板语言中有…...

编程日记 2025/6/2 7:39:48

【JavaWeb后端学习笔记】MySQL的数据操作语言（Data Manipulation Language，DML）

MySQL DML 0、准备表结构1、添加数据1.1 指定字段添加数据&#xff08;单条&#xff09;1.2 全部字段添加数据&#xff08;单条&#xff09;1.3 指定字段添加数据&#xff08;批量&#xff09;1.4 全部字段添加数据&#xff08;批量&#xff09; 2、修改数据3、删除数据 MySQL的…...

编程日记 2025/6/2 8:29:54

【SpringBoot】Day10-09 动态SQL-XML文件

动态SQL-XML文件的三点规范 1.XML映射文件的名称与Mapper接口名称保持一致&#xff0c;并且将XML映射文件和Mapper接口放置在相同包下&#xff08;同包同名&#xff09;- 在项目开发当中&#xff0c;一般都是一个接口对应一份儿映射配置文件&#xff1b; 2.XML映射文件的namesp…...

编程日记 2025/6/3 4:27:21

Linux絮絮叨(三) Ubuntu桌面版添加中文拼音输入法

步骤很详细&#xff0c;直接上教程 一. 配置安装简体拼音输入法 #安装相应的平台支持包 sudo apt install ibus-gtk ibus-gtk3# 安装简体拼音输入法 sudo apt install ibus-pinyin安装完成如果下面的步骤找不到对应输入法可以重启一下&#xff0c;一般不需要 二. 添加简体拼音…...

编程日记 2025/6/4 6:29:48

rockit 学习、开发笔记(六)(VENC)

前言 上节我们讲到了VDEC解码模块&#xff0c;那当然少不了VENC编码模块了&#xff0c;一般有编解码的需求都是为了压缩视频的大小&#xff0c;方便减少传输所占用的带宽。 概述 VENC 模块&#xff0c;即视频编码模块。本模块支持多路实时编码&#xff0c;且每路编码独立&am…...

编程日记 2025/6/2 6:55:24

Python+Django框架山东济南景点数据可视化系统网站作品截图和开题报告参考

博主介绍&#xff1a;黄菊华老师《Vue.js入门与商城开发实战》《微信小程序商城开发》图书作者&#xff0c;CSDN博客专家&#xff0c;在线教育专家&#xff0c;CSDN钻石讲师&#xff1b;专注大学生毕业设计教育、辅导。 所有项目都配有从入门到精通的基础知识视频课程&#xff…...

编程日记 2025/6/1 10:34:23

【C#】键值对的一种常见数据结构Dictionary＜TKey, TValue＞

在 C# 中&#xff0c;Dictionary<TKey, TValue> 是一个 键值对&#xff08;key-value&#xff09;集合&#xff0c;是一种非常常见的数据结构。它允许通过 键&#xff08;key&#xff09;来快速查找与之相关的 值&#xff08;value&#xff09;。你可以将其类比为一个映射…...

编程日记 2025/6/4 1:16:09

SQL Server：调用的目标发生了异常。（mscorlib）

我之前安装的SQL Server是2014版本&#xff0c;SSMS运行也很流畅&#xff0c;有一次有个同事让我链接云服务器SQL地址&#xff0c;直接报上图的错误&#xff0c;把我弄的一愣一愣的。 后面才发现&#xff0c;这是版本太低导致的&#xff0c;但是你如果使用Navicat是没有问题的…...

编程日记 2025/5/27 15:28:03

windows 上ffmpeg编译好的版本选择

1. Gyan.dev Gyan.dev 是一个广受信赖的 FFmpeg 预编译库提供者&#xff0c;提供多种版本的 FFmpeg&#xff0c;包括静态和动态链接版本。 下载链接: https://www.gyan.dev/ffmpeg/builds/ 特点&#xff1a; 提供最新稳定版和开发版。 支持静态和共享&#xff08;动态&…...

编程日记 2025/6/4 17:21:18

前端工程化面试题（一）

如何使用 Docker 部署前端项目&#xff1f; 使用 Docker 部署前端项目通常涉及以下几个步骤&#xff1a; 创建项目&#xff1a;首先&#xff0c;需要在本地创建并配置好前端项目。 准备 Docker 文件&#xff1a; .dockerignore&#xff1a;这个文件用于排除不需要上传到 Dock…...

编程日记 2025/6/2 12:03:56

Java设计模式笔记（二）

十四、模版方法模式 1、介绍 1&#xff09;模板方法模式(Template Method Pattern)&#xff0c;又叫模板模式(Template Patern)&#xff0c;在一个抽象类公开定义了执行它的方法的模板。它的子类可以按需重写方法实现&#xff0c;但调用将以抽象类中定义的方式进行。 2&…...

编程日记 2025/6/3 6:35:19

vscode（一）安装（ubuntu20.04）

1、更新软件包列表 sudo apt update2、安装依赖包 sudo apt install software-properties-common apt-transport-https wget3、导入Microsoft GPG密钥 wget -q https://packages.microsoft.com/keys/microsoft.asc -O- | sudo apt-key add -4、向系统添加VSCode存储库 sudo…...

编程日记 2025/5/29 1:28:57

C# 命名空间（Namespace）

文章目录 前言一、命名空间的定义与使用基础&#xff08;一&#xff09;定义语法与规则&#xff08;二&#xff09;调用命名空间内元素 二、using 关键字三、嵌套命名空间 前言 命名空间&#xff08;Namespace&#xff09;在于提供一种清晰、高效的方式&#xff0c;将一组名称与…...

编程日记 2025/5/30 23:27:51

云计算介绍_3（计算虚拟化——cpu虚拟化、内存虚拟化、io虚拟化、常见集群策略、华为FC）

计算虚拟化 1.计算虚拟化介绍1.1 计算虚拟化 分类&#xff08;cpu虚拟化、内存虚拟化、IO虚拟化&#xff09;1.2 cpu虚拟化1.3 内存虚拟化1.4 IO虚拟化1.5 常见的集群的策略1.6 华为FC 1.计算虚拟化介绍 1.1 计算虚拟化 分类&#xff08;cpu虚拟化、内存虚拟化、IO虚拟化&#…...

编程日记 2025/6/3 3:26:29

Flutter开发App，编译条件下UI没问题，打包后UI出现问题

刚入门Flutter3个月&#xff0c;终于进入项目打包阶段&#xff0c;发现之前编译环境下都正常的UI&#xff0c;忽然有小部分出现异常&#xff0c;不显示UI部分了。查了2个小时都没发现原因。因为编译环境下&#xff0c;Android、iOS端都正常显示。但是进入打包安装后&#xff0c…...

编程日记 2025/6/2 6:34:37

Python+OpenCV系列：Python和OpenCV的结合和发展

PythonOpenCV系列&#xff1a;Python和OpenCV的结合和发展 **引言****Python语言的发展****1.1 Python的诞生与发展****1.2 Python的核心特性与优势****1.3 Python的应用领域** **OpenCV的发展****2.1 OpenCV的起源与发展****2.2 OpenCV的功能特性****2.3 OpenCV的应用场景** *…...

编程日记 2025/6/3 3:50:29

报错 JSON.parse: expected property name or ‘}‘，JSON数据中对象的key值不为字符串

报错 JSON.parse: expected property name or ‘}’ 原因 多是因为数据转换时出错&#xff0c;可能是存在单引号或者对象key值不为string导致 这里记录下我遇见的问题&#xff08;后端给的JSON数据里&#xff0c;对象key值不为string&#xff09; 现在后端转换JSON数据大多…...

编程日记 2025/6/3 13:15:12

Flutter：商品多规格内容总结，响应式数据，高亮切换显示。

如图所示&#xff1a; 代码为练习时写的项目&#xff0c;写的一般&#xff0c;功能实现了&#xff0c;等以后再来优化。 自己模拟的数据结构 var data {id:1,name:精品小米等多种五谷杂粮精品小等多种五谷杂粮,logo:https://cdn.uviewui.com/uview/swiper/1.jpg,price:100.5…...

编程日记 2025/6/2 9:40:29

WPF+LibVLC开发播放器-LibVLC播放控制

接上一篇&#xff1a; LibVLC在C#中的使用 实现LibVLC播放器播放控制 界面 界面上添加一个Button按钮用于控制播放 <ButtonGrid.Row"1"Width"88"Height"24"Margin"10,0,0,0"HorizontalAlignment"Left"VerticalAlignme…...

编程日记 2025/5/30 16:56:08

Mac环境下brew安装LNMP

安装不同版本PHP 在Mac环境下同时运行多个版本的PHP&#xff0c;同Linux环境一样&#xff0c;都是将后台运行的php-fpm设置为不同的端口号&#xff0c;下面将已php7.2 和 php7.4为例 添加 tap 目的&#xff1a;homebrew仅保留最近的php版本&#xff0c;可能没有你需要的版本…...

编程日记 2025/6/3 1:48:59

nodejs后端项目使用pm2部署

nodejs后端项目使用pm2部署 安装pm2 npm install pm2 -g查看版本号 pm2 --version启动项目 pm2 start app.js# 设置别名 pm2 start app.js --name demo停止项目 pm2 stop [AppName] pm2 stop [ID]# 停止所有项目 pm2 stop all重启项目 pm2 restart [AppName] pm2 re…...

编程日记 2025/6/4 1:54:10

【C++】深入理解字符变量取地址的特殊性与内存管理机制详解

博客主页&#xff1a; [小ᶻ☡꙳ᵃⁱᵍᶜ꙳] 本文专栏: C 文章目录 &#x1f4af;前言&#x1f4af;栈内存中的变量分配&#xff1a;谁先谁后&#xff1f;cout 的输出行为&#xff1a;按顺序执行&#xff0c;按地址递增读取代码执行顺序与内存布局的关系编译器优化的影响 &…...

编程日记 2025/6/4 12:22:18

【银河麒麟操作系统真实案例分享】内存黑洞导致服务器卡死分析全过程

了解更多银河麒麟操作系统全新产品&#xff0c;请点击访问 麒麟软件产品专区&#xff1a;https://product.kylinos.cn 开发者专区&#xff1a;https://developer.kylinos.cn 文档中心&#xff1a;https://documentkylinos.cn 现象描述 机房显示器连接服务器后黑屏&#xff…...

编程日记 2025/6/4 11:35:10

蓝桥杯软件赛系列---lesson1

&#x1f308;个人主页&#xff1a;羽晨同学 &#x1f4ab;个人格言:“成为自己未来的主人~” 我们今天会再开一个系列&#xff0c;那就是蓝桥杯系列&#xff0c;我们会从最基础的开始讲起&#xff0c;大家想要备战明年蓝桥杯的&#xff0c;让我们一起加油。 工具安装 DevC…...

编程日记 2025/6/1 0:44:11

工业异常检测-CVPR2024-新的3D异常数据合成办法和自监督网络IMRNet

论文&#xff1a;https://arxiv.org/pdf/2311.14897v3.pdf 项目&#xff1a;https://github.com/chopper-233/anomaly-shapenet 这篇论文主要关注的是3D异常检测和定位&#xff0c;这是一个在工业质量检查中至关重要的任务。作者们提出了一种新的方法来合成3D异常数据&#x…...

编程日记 2025/5/31 10:55:29

HeidiSQL：MySQL图形化管理工具深度解析

本文还有配套的精品资源&#xff0c;点击获取 简介&#xff1a;HeidiSQL是一款开源的MySQL图形化管理工具&#xff0c;适用于多种数据库系统&#xff0c;如MySQL、MariaDB、SQL Server、PostgreSQL和SQLite。其提供的直观操作界面和丰富的功能简化了数据库操作&#xff0c;包…...

编程日记 2025/6/1 9:20:57

【Redis】深入解析Redis缓存机制：全面掌握缓存更新、穿透、雪崩与击穿的终极指南

文章目录 一、Redis缓存机制概述1.1 Redis缓存的基本原理1.2 常见的Redis缓存应用场景 二、缓存更新机制2.1 缓存更新的策略2.2 示例代码&#xff1a;主动更新缓存 三、缓存穿透3.1 缓存穿透的原因3.2 缓解缓存穿透的方法3.3 示例代码&#xff1a;使用布隆过滤器 四、缓存雪崩4…...

编程日记 2025/6/2 0:39:07

Flask使用Celery与多进程管理：优雅处理长时间任务与子进程终止技巧(multiprocessing)(subprocess)

在许多任务处理系统中&#xff0c;我们需要使用异步任务队列来处理繁重的计算或长时间运行的任务&#xff0c;如模型训练。Celery是一个广泛使用的分布式任务队列&#xff0c;而在某些任务中&#xff0c;尤其是涉及到调用独立脚本的场景中&#xff0c;我们需要混合使用multipro…...

编程日记 2025/6/4 1:54:09

【PyTorch】torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

报错说明 torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 报错如图所示 报错分析 该报错是 torch 和 CUDA 版本不兼容导致。 &#xff08;一般N卡自带的CUDA版本与最新的torch版本相差较大&#xff09; 解决方案 1.查看自己的CUDA版本 # 查看自己的…...

编程日记 2025/5/30 16:09:09

使用android studio写一个Android的远程通信软件（APP）,有通讯的发送和接收消息界面

以下是使用 Android Studio 基于 Java 语言编写一个简单的 Android APP 实现远程通信&#xff08;这里以 TCP 通信为例&#xff09;的代码示例&#xff0c;包含基本的通信界面以及发送和接收消息功能。 1. 创建项目 打开 Android Studio&#xff0c;新建一个 Empty Activity …...

编程日记 2025/6/4 4:25:45

突破空间限制！从2D到3D：北大等开源Lift3D，助力精准具身智能操作！

文章链接&#xff1a;https://arxiv.org/pdf/2411.18623 项目链接&#xff1a;https://lift3d-web.github.io/ 亮点直击 提出了Lift3D&#xff0c;通过系统地提升隐式和显式的3D机器人表示&#xff0c;提升2D基础模型&#xff0c;构建一个3D操作策略。 对于隐式3D机器人表示&a…...

编程日记 2025/6/4 13:54:14

Android KEY的哪些事儿

目录 一、APK应用签名 1、什么是APK应用签名&#xff1f; 1.1 目的和作用&#xff1f; 1.2 长什么样子&#xff1f; 2、APK应用签名使用流程 步骤一&#xff1a;如何生成APK应用签名文件&#xff1f; 步骤二&#xff1a;如何集成APK应用签名文件&#xff1f; 步骤三&am…...

编程日记 2025/5/30 8:04:35

李宏毅深度学习-Pytorch Tutorial2

什么是张量&#xff1f; 张量&#xff08;Tensor&#xff09;是深度学习和机器学习中一个非常基础且重要的概念。在数学上&#xff0c;张量可以被看作是向量和矩阵的泛化。简单来说&#xff0c;张量是一种多维数组&#xff0c;它可以表示标量&#xff08;0维&#xff09;、向量…...

编程日记 2025/6/3 9:06:38

【译】为 SAP 表维护视图 (SM30) 创建选择屏幕

原文标题&#xff1a;Create Selection Screen for SAP Table Maintenance View (SM30) 原文链接&#xff1a; https://www.saphub.com/abap-dictionary/sap-abap-tmg-selection-screen/ 通常&#xff0c;带有单个屏幕的 SAP 表维护视图 (SM30) 会显示表中的所有记录&#xff…...

编程日记 2025/5/30 20:33:58

element Plus中 el-table表头宽度自适应，不换行

在工作中&#xff0c;使用el-table表格进行开发后&#xff0c;遇到了小屏幕显示器上显示表头文字会出现换行展示&#xff0c;比较影响美观&#xff0c;因此需要让表头的宽度变为不换行&#xff0c;且由内容自动撑开。 以下是作为工作记录&#xff0c;用于demo演示教程 先贴个…...

编程日记 2025/6/1 16:31:00

C语言程序设计P5-4【应用函数进行程序设计 | 第四节】——知识要点：数组作函数参数

知识要点&#xff1a;数组作函数参数 视频&#xff1a; 目录 一、任务分析 二、必备知识与理论 三、任务实施 一、任务分析 任务要求用选择法对数组中的 10 个整数按由小到大的顺序排序&#xff0c;前面在讲解数组时讲冒泡法排序曾提到选择法排序的思想。 所谓选择法就是…...

编程日记 2025/6/4 17:48:37

时间序列模型在LSTM中的特征输入

这里写目录标题 前言LSTM的输入组成时间步例子 实际代码解读特征提取处理成dataloader格式&#xff08;用于输入到模型当中&#xff09;对应到lstm的模型创建代码 总结 前言 本文章将帮助理解如何将一个时间序列的各种特征&#xff08;年月日的时间特征&#xff0c;滚动窗口滞…...

编程日记 2025/6/2 3:42:23

Python_Flask02

所有人都不许学Java了&#xff0c;都来学Python&#xff01; 如果不来学的话请网爆我的老师 连接前的准备 安装pymysql 和 flask_sqlalchemy&#xff0c;安装第三下面两个所需要的包才能连接上数据库 pip install pymysql pip install flask_sqlalchemy pymysql是一个Pyth…...

编程日记 2025/5/30 0:12:36

threejs相机辅助对象cameraHelper

为指定相机创建一个辅助对象&#xff0c;显示这个相机的视锥。 想要在场景里面显示相机的视锥&#xff0c;需要创建两个相机。 举个例子&#xff0c;场景中有个相机A&#xff0c;想要显示相机A的视锥&#xff0c;那么需要一个相机B&#xff0c;把B放在A的后面&#xff0c;两个…...

编程日记 2025/6/4 12:35:45

断点续传+测试方法完整示例

因为看不懂网上的断点续传案例&#xff0c;而且又不能直接复制使用&#xff0c;干脆自己想想写了一个。 上传入参类&#xff1a; import com.fasterxml.jackson.annotation.JsonIgnore; import io.swagger.annotations.ApiModel; import io.swagger.annotations.ApiModelProp…...

编程日记 2025/5/30 15:47:25

C#设计模式--状态模式（State Pattern）

状态模式是一种行为设计模式&#xff0c;它允许对象在其内部状态发生变化时改变其行为。这种模式的核心思想是将状态封装在独立的对象中&#xff0c;而不是将状态逻辑散布在整个程序中。 用途 简化复杂的条件逻辑&#xff1a;通过将不同的状态封装在不同的类中&#xff0c;可…...

编程日记 2025/5/30 16:38:42

Excel技巧：如何批量调整excel表格中的图片？

插入到excel表格中的图片大小不一&#xff0c;如何做到每张图片都完美的与单元格大小相同&#xff1f;并且能够根据单元格来改变大小&#xff1f;今天分享&#xff0c;excel表格里的图片如何批量调整大小。 方法如下&#xff1a; 点击表格中的一个图片&#xff0c;然后按住Ct…...

编程日记 2025/6/3 13:11:15

hadoop中导出表与数据的步骤

大家好&#xff0c;我是 V 哥。在Hadoop中导出表与数据&#xff0c;可以通过多种方式实现&#xff0c;包括使用Hive的EXPORT命令、MapReduce作业、Hive查询以及Sqoop工具。下面V 哥将详细介绍这些步骤和一些代码示例&#xff0c;来帮助大家更好理解。 1. 使用Hive的EXPORT命令…...

编程日记 2025/6/2 22:51:26

springBoot中的日志级别在哪里配置

在Spring Boot中&#xff0c;日志级别的配置可以通过多种方式来实现&#xff0c;主要包括在配置文件中设置、使用自定义的logback配置文件&#xff0c;以及在代码中动态配置等。以下是一些具体的配置方法&#xff1a; 一、在配置文件中设置日志级别 Spring Boot默认使用appli…...

编程日记 2025/6/3 10:51:28

17. Threejs案例-Three.js创建多个立方体

17. Threejs案例-Three.js创建多个立方体 实现效果 知识点 WebGLRenderer (WebGL渲染器) WebGLRenderer 是 Three.js 中用于渲染 WebGL 场景的核心类。它负责将场景中的对象渲染到画布上。 构造器 new THREE.WebGLRenderer(parameters) 参数类型描述parametersObject可选…...

编程日记 2025/6/4 0:19:41

数据结构——有序二叉树的删除

在上一篇博客中&#xff0c;我们介绍了有序二叉树的构建、遍历、查找。 数据结构——有序二叉树的构建&遍历&查找-CSDN博客文章浏览阅读707次&#xff0c;点赞18次&#xff0c;收藏6次。因为数据的类型决定数据在内存中的存储形式。left right示意为左右节点其类型也为…...

编程日记 2025/5/31 18:21:59

力扣1401. 圆和矩形是否有重叠

用矢量计算&#xff1a; class Solution { public:bool checkOverlap(int radius, int xCenter, int yCenter, int x1, int y1, int x2, int y2) {//矩形中心float Tx(float)(x1x2)/2;float Ty(float)(y1y2)/2;//强行进行对称操作&#xff0c;只考虑第一象限if(xCenter<Tx)…...

编程日记 2025/6/1 18:07:04

idea连接到docker出现 org.apache.hc.client5.http.ConnectTimeoutException 异常怎么办？

前情提要 我电脑是win11&#xff0c;我安装了centOS7虚拟机&#xff0c;配置linux环境 idea是2024社区免费版本 我就这一步步排查问题&#xff0c;终于发现了是因为我的2375端口没有ipv4开放&#xff0c;只在ipv6开放 踩坑提醒&#xff1a; 对了&#xff0c;一个一个问题排…...

编程日记 2025/5/31 15:04:04

一番赏小程序定制开发，打造全新抽赏体验平台

随着盲盒的热潮来袭&#xff0c;作为传统的潮玩方式一番赏也再次受到了大家的关注&#xff0c;市场热度不断上升&#xff01; 一番赏能够让玩家百分百中奖&#xff0c;商品种类丰富、收藏价值高&#xff0c;拥有各种IP&#xff0c;从而吸引着各个圈子的粉丝玩家&#xff0c;用…...

编程日记 2025/5/30 12:02:30