深度學習項目--基于DenseNet網絡的“乳腺癌圖像識別”，準確率90%+，pytorch復現

  • 🍨 本文為🔗365天深度學習訓練營 中的學習記錄博客
  • 🍖 原作者：K同學啊

前言

  • 如果說最經典的神經網絡，ResNet肯定是一個，從ResNet發布后，很多人做了修改，denseNet網絡無疑是最成功的一個，它采用密集型連接，將通道數連接在一起；
  • 本文是基于上一篇復現DenseNet121模型，做一個乳腺癌圖像識別，效果還行，準確率0.9+;
  • CNN經典網絡之“DenseNet”簡介，源碼研究與復現(pytorch)： https://blog.csdn.net/weixin_74085818/article/details/146102290?spm=1001.2014.3001.5501
  • 歡迎收藏 + 關注，本人將會持續更新

文章目錄

    • 1、導入數據
      • 1、導入庫
      • 2、查看數據信息和導入數據
      • 3、展示數據
      • 4、數據導入
      • 5、數據劃分
      • 6、動態加載數據
    • 2、構建DenseNet121網絡
    • 3、模型訓練
      • 1、構建訓練集
      • 2、構建測試集
      • 3、設置超參數
    • 4、模型訓練
    • 5、結果可視化
    • 6、模型評估

1、導入數據

1、導入庫

import torch  
import torch.nn as nn
import torchvision 
import numpy as np 
import os, PIL, pathlib 
from collections import OrderedDict
import re
from torch.hub import load_state_dict_from_url# 設置設備
device = "cuda" if torch.cuda.is_available() else "cpu"device 

'cuda'

2、查看數據信息和導入數據

data_dir = "./data/"data_dir = pathlib.Path(data_dir)# 類別數量
classnames = [str(path).split("\\")[0] for path in os.listdir(data_dir)]classnames

['0', '1']

3、展示數據

import matplotlib.pylab as plt  
from PIL import Image # 獲取文件名稱
data_path_name = "./data/0/"  # 不患病的
data_path_list = [f for f in os.listdir(data_path_name) if f.endswith(('jpg', 'png'))]# 創建畫板
fig, axes = plt.subplots(2, 8, figsize=(16, 6))for ax, img_file in zip(axes.flat, data_path_list):path_name = os.path.join(data_path_name, img_file)img = Image.open(path_name) # 打開# 顯示ax.imshow(img)ax.axis('off')plt.show()

?
在這里插入圖片描述

?

4、數據導入

from torchvision import transforms, datasets # 數據統一格式
img_height = 224
img_width = 224 data_tranforms = transforms.Compose([transforms.Resize([img_height, img_width]),transforms.ToTensor(),transforms.Normalize(   # 歸一化mean=[0.485, 0.456, 0.406],std=[0.229, 0.224, 0.225] )
])# 加載所有數據
total_data = datasets.ImageFolder(root="./data/", transform=data_tranforms)

5、數據劃分

# 大小 8 : 2
train_size = int(len(total_data) * 0.8)
test_size = len(total_data) - train_size train_data, test_data = torch.utils.data.random_split(total_data, [train_size, test_size])

6、動態加載數據

batch_size = 64train_dl = torch.utils.data.DataLoader(train_data,batch_size=batch_size,shuffle=True
)test_dl = torch.utils.data.DataLoader(test_data,batch_size=batch_size,shuffle=False
)

# 查看數據維度
for data, labels in train_dl:print("data shape[N, C, H, W]: ", data.shape)print("labels: ", labels)break

data shape[N, C, H, W]:  torch.Size([64, 3, 224, 224])
labels:  tensor([1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1,1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0,1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 0, 1])

2、構建DenseNet121網絡

import torch.nn.functional as F# 實現DenseBlock中的部件：DenseLayer
'''  
1、BN + ReLU: 處理部分，首先進行歸一化，然后在用激活函數ReLU
2、Bottlenck Layer：稱為瓶頸層，這個層在yolov5中常用，但是yolov5中主要用于特征提取+維度降維，這里采用1 * 1卷積核 + 3 * 3的卷積核進行卷積操作，目的：減少輸入輸入特征維度
3、BN + ReLU：對 瓶頸層 數據進行歸一化，ReLU激活函數，歸一化可以確保梯度下降的時候較為平穩
4、3 * 3 生成新的特征圖
'''
class _DenseLayer(nn.Sequential):def __init__(self, num_input_features, growth_rate, bn_size, drop_rate):'''  num_input_features: 輸入特征數，也就是通道數，在DenseNet中，每一層都會接受之前層的輸出作為輸入，故，這個數值通常會隨著網絡深度增加而增加growth_rate: 增長率，這個是 DenseNet的核心概念，決定了每一層為全局狀態貢獻的特征數量，他的用處主要在于決定了中間瓶頸層的輸出通道，需要結合代碼去研究bn_size: 瓶頸層中輸出通道大小，含義：在使用1 * 1卷積核去提取特征數時，目標通道需要擴展到growth_rate的多少倍倍數， bn_size * growth_rate(輸出維度)drop_rate: 使用Dropout的參數'''super(_DenseLayer, self).__init__()self.add_module("norm1", nn.BatchNorm2d(num_input_features))self.add_module("relu1", nn.ReLU(inplace=True))# 輸出維度： bn_size * growth_rate, 1 * 1卷積核，步伐為1，只起到特征提取作用self.add_module("conv1", nn.Conv2d(num_input_features, bn_size * growth_rate, stride=1, kernel_size=1, bias=False))self.add_module("norm2", nn.BatchNorm2d(bn_size * growth_rate))self.add_module("relu2", nn.ReLU(inplace=True))# 輸出通道：growth_rate, 維度計算：不變self.add_module("conv2", nn.Conv2d(bn_size * growth_rate, growth_rate, stride=1, kernel_size=3, padding=1, bias=False))self.drop_rate = drop_ratedef forward(self, x):new_features = super(_DenseLayer, self).forward(x)  # 傳播if self.drop_rate > 0:new_features = F.dropout(new_features, p=self.drop_rate, training=self.training)  # self.training 繼承nn.Sequential，是否訓練模式# 模型融合，即，特征通道融合，形成新的特征圖return torch.cat([x, new_features], dim=1)  # (N, C, H, W)  # 即 C1 + C2，通道上融合'''  
DenseNet網絡核心由DenseBlock模塊組成，DenseBlock網絡由DenseLayer組成，從 DenseLayer 可以看出，DenseBlock是密集連接，每一層的輸入不僅包含前一層的輸出，還包含網絡中所有之前層的輸出
'''
# 構建DenseBlock模塊, 通過上圖
class _DenseBlock(nn.Sequential):# num_layers 幾層DenseLayer模塊def __init__(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate):super(_DenseBlock, self).__init__()for i in range(num_layers):layer = _DenseLayer(num_input_features + i * growth_rate, growth_rate, bn_size, drop_rate)self.add_module("denselayer%d" % (i + 1), layer)# Transition層，用于維度壓縮# 組成：一個卷積層 + 一個池化層
class _Transition(nn.Sequential):def __init__(self, num_init_features, num_out_features):super(_Transition, self).__init__()self.add_module("norm", nn.BatchNorm2d(num_init_features))self.add_module("relu", nn.ReLU(inplace=True))self.add_module("conv", nn.Conv2d(num_init_features, num_out_features, kernel_size=1, stride=1, bias=False))# 降維self.add_module("pool", nn.AvgPool2d(2, stride=2))# 搭建DenseNet網絡
class DenseNet(nn.Module):def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16), num_init_features=64, bn_size=4, compression_rate=0.5, drop_rate=0.5, num_classes=1000):'''  growth_rate、num_init_features、num_init_features、drop_rate 和denselayer一樣block_config : 參數在 DenseNet 架構中用于指定每個 Dense Block 中包含的層數, 如：DenseNet-121: block_config=(6, 12, 24, 16) 表示第一個 Dense Block 包含 6 層，第二個包含 12 層，第三個包含 24 層，第四個包含 16 層。DenseNet-169: block_config=(6, 12, 32, 32)DenseNet-201: block_config=(6, 12, 48, 32)DenseNet-264: block_config=(6, 12, 64, 48)compression_rate: 壓縮維度, DenseNet 中用于 Transition Layer（過渡層）的一個重要參數，它控制了從一個 Dense Block 到下一個 Dense Block 之間特征維度的壓縮程度'''super(DenseNet, self).__init__()# 第一層卷積# OrderedDict，讓模型層有序排列self.features = nn.Sequential(OrderedDict([# 輸出維度：((w - k + 2 * p) / s) + 1("conv0", nn.Conv2d(3, num_init_features, kernel_size=7, stride=2, padding=3, bias=False)),("norm0", nn.BatchNorm2d(num_init_features)),("relu0", nn.ReLU(inplace=True)),("pool0", nn.MaxPool2d(3, stride=2, padding=1))  # 降維]))# 搭建DenseBlock層num_features = num_init_features# num_layers: 層數for i, num_layers in enumerate(block_config):block = _DenseBlock(num_layers, num_features, bn_size, growth_rate, drop_rate)# nn.Module 中features封裝了nn.Sequentialself.features.add_module("denseblock%d" % (i + 1), block)'''  # 這個計算反映了 DenseNet 中的一個關鍵特性：每一層輸出的特征圖（即新增加的通道數）由 growth_rate 決定，# 并且這些新生成的特征圖會被傳遞給該 Dense Block 中的所有后續層以及下一個 Dense Block。'''num_features += num_layers * growth_rate  # 疊加，每一次疊加# 判斷是否需要使用Transition層if i != len(block_config) - 1:transition = _Transition(num_features, int(num_features*compression_rate)) # compression_rate 作用self.features.add_module("transition%d" % (i + 1), transition)num_features = int(num_features*compression_rate)  # 更新維度# 最后一層self.features.add_module("norm5", nn.BatchNorm2d(num_features))self.features.add_module("relu5", nn.ReLU(inplace=True))# 分類層         self.classifier = nn.Linear(num_features, num_classes)# params initialization         for m in self.modules():             if isinstance(m, nn.Conv2d):         '''如果當前模塊是一個二維卷積層 (nn.Conv2d)，那么它的權重 (m.weight) 將通過 Kaiming 正態分布 (kaiming_normal_) 進行初始化。這種初始化方式特別適合與ReLU激活函數一起使用，有助于緩解深度網絡中的梯度消失問題，促進有效的訓練。  '''       nn.init.kaiming_normal_(m.weight)             elif isinstance(m, nn.BatchNorm2d):      '''  對于二維批歸一化層 (nn.BatchNorm2d)，偏置項 (m.bias) 被初始化為0，而尺度因子 (m.weight) 被初始化為1。這意味著在沒有數據經過的情況下，批歸一化層不會對輸入進行額外的縮放或偏移，保持輸入不變。'''           nn.init.constant_(m.bias, 0)                 nn.init.constant_(m.weight, 1)             elif isinstance(m, nn.Linear):        '''  對于全連接層 (nn.Linear)，只對其偏置項 (m.bias) 進行了初始化，設置為0'''         nn.init.constant_(m.bias, 0)def forward(self, x):features = self.features(x)out = F.avg_pool2d(features, 7, stride=1).view(x.size(0), -1)out = self.classifier(out)return outmodel = DenseNet(num_init_features=64, growth_rate=32, block_config=(6, 12, 12, 16))model.to(device)

DenseNet((features): Sequential((conv0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)(norm0): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu0): ReLU(inplace=True)(pool0): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)(denseblock1): _DenseBlock((denselayer1): _DenseLayer((norm1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer2): _DenseLayer((norm1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(96, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer3): _DenseLayer((norm1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer4): _DenseLayer((norm1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(160, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer5): _DenseLayer((norm1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(192, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer6): _DenseLayer((norm1): BatchNorm2d(224, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(224, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)))(transition1): _Transition((norm): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(pool): AvgPool2d(kernel_size=2, stride=2, padding=0))(denseblock2): _DenseBlock((denselayer1): _DenseLayer((norm1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer2): _DenseLayer((norm1): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(160, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer3): _DenseLayer((norm1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(192, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer4): _DenseLayer((norm1): BatchNorm2d(224, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(224, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer5): _DenseLayer((norm1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer6): _DenseLayer((norm1): BatchNorm2d(288, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(288, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer7): _DenseLayer((norm1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(320, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer8): _DenseLayer((norm1): BatchNorm2d(352, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(352, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer9): _DenseLayer((norm1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer10): _DenseLayer((norm1): BatchNorm2d(416, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(416, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer11): _DenseLayer((norm1): BatchNorm2d(448, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(448, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer12): _DenseLayer((norm1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(480, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)))(transition2): _Transition((norm): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)(pool): AvgPool2d(kernel_size=2, stride=2, padding=0))(denseblock3): _DenseBlock((denselayer1): _DenseLayer((norm1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer2): _DenseLayer((norm1): BatchNorm2d(288, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(288, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer3): _DenseLayer((norm1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(320, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer4): _DenseLayer((norm1): BatchNorm2d(352, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(352, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer5): _DenseLayer((norm1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer6): _DenseLayer((norm1): BatchNorm2d(416, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(416, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer7): _DenseLayer((norm1): BatchNorm2d(448, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(448, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer8): _DenseLayer((norm1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(480, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer9): _DenseLayer((norm1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer10): _DenseLayer((norm1): BatchNorm2d(544, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(544, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer11): _DenseLayer((norm1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(576, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer12): _DenseLayer((norm1): BatchNorm2d(608, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(608, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)))(transition3): _Transition((norm): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu): ReLU(inplace=True)(conv): Conv2d(640, 320, kernel_size=(1, 1), stride=(1, 1), bias=False)(pool): AvgPool2d(kernel_size=2, stride=2, padding=0))(denseblock4): _DenseBlock((denselayer1): _DenseLayer((norm1): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(320, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer2): _DenseLayer((norm1): BatchNorm2d(352, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(352, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer3): _DenseLayer((norm1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(384, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer4): _DenseLayer((norm1): BatchNorm2d(416, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(416, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer5): _DenseLayer((norm1): BatchNorm2d(448, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(448, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer6): _DenseLayer((norm1): BatchNorm2d(480, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(480, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer7): _DenseLayer((norm1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer8): _DenseLayer((norm1): BatchNorm2d(544, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(544, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer9): _DenseLayer((norm1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(576, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer10): _DenseLayer((norm1): BatchNorm2d(608, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(608, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer11): _DenseLayer((norm1): BatchNorm2d(640, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(640, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer12): _DenseLayer((norm1): BatchNorm2d(672, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(672, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer13): _DenseLayer((norm1): BatchNorm2d(704, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(704, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer14): _DenseLayer((norm1): BatchNorm2d(736, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(736, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer15): _DenseLayer((norm1): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(768, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False))(denselayer16): _DenseLayer((norm1): BatchNorm2d(800, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu1): ReLU(inplace=True)(conv1): Conv2d(800, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)(norm2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu2): ReLU(inplace=True)(conv2): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)))(norm5): BatchNorm2d(832, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)(relu5): ReLU(inplace=True))(classifier): Linear(in_features=832, out_features=1000, bias=True)
)

3、模型訓練

1、構建訓練集

def train(dataloader, model, loss_fn, optimizer):size = len(dataloader.dataset)batch_size = len(dataloader)train_acc, train_loss = 0, 0 for X, y in dataloader:X, y = X.to(device), y.to(device)# 訓練pred = model(X)loss = loss_fn(pred, y)# 梯度下降法optimizer.zero_grad()loss.backward()optimizer.step()# 記錄train_loss += loss.item()train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()train_acc /= sizetrain_loss /= batch_sizereturn train_acc, train_loss

2、構建測試集

def test(dataloader, model, loss_fn):size = len(dataloader.dataset)batch_size = len(dataloader)test_acc, test_loss = 0, 0 with torch.no_grad():for X, y in dataloader:X, y = X.to(device), y.to(device)pred = model(X)loss = loss_fn(pred, y)test_loss += loss.item()test_acc += (pred.argmax(1) == y).type(torch.float).sum().item()test_acc /= sizetest_loss /= batch_sizereturn test_acc, test_loss

3、設置超參數

loss_fn = nn.CrossEntropyLoss()  # 損失函數     
learn_lr = 1e-4             # 超參數
optimizer = torch.optim.Adam(model.parameters(), lr=learn_lr)   # 優化器

4、模型訓練

通過實驗發現，還是設置20輪次附件最好

import copytrain_acc = []
train_loss = []
test_acc = []
test_loss = []epoches = 20best_acc = 0    # 設置一個最佳準確率，作為最佳模型的判別指標for i in range(epoches):model.train()epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)model.eval()epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)# 保存最佳模型到 best_model     if epoch_test_acc > best_acc:         best_acc   = epoch_test_acc         best_model = copy.deepcopy(model)train_acc.append(epoch_train_acc)train_loss.append(epoch_train_loss)test_acc.append(epoch_test_acc)test_loss.append(epoch_test_loss)# 獲取當前的學習率     lr = optimizer.state_dict()['param_groups'][0]['lr']# 輸出template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}')print(template.format(i + 1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss))PATH = './best_model.pth'  # 保存的參數文件名 
torch.save(best_model.state_dict(), PATH)print("Done")

Epoch: 1, Train_acc:79.3%, Train_loss:1.948, Test_acc:84.6%, Test_loss:1.079
Epoch: 2, Train_acc:85.3%, Train_loss:0.395, Test_acc:85.2%, Test_loss:0.721
Epoch: 3, Train_acc:87.3%, Train_loss:0.318, Test_acc:86.5%, Test_loss:0.526
Epoch: 4, Train_acc:89.0%, Train_loss:0.277, Test_acc:86.6%, Test_loss:0.494
Epoch: 5, Train_acc:89.0%, Train_loss:0.266, Test_acc:87.9%, Test_loss:0.400
Epoch: 6, Train_acc:89.6%, Train_loss:0.252, Test_acc:84.6%, Test_loss:0.524
Epoch: 7, Train_acc:90.3%, Train_loss:0.239, Test_acc:85.5%, Test_loss:0.445
Epoch: 8, Train_acc:90.2%, Train_loss:0.235, Test_acc:87.6%, Test_loss:0.359
Epoch: 9, Train_acc:90.0%, Train_loss:0.235, Test_acc:89.3%, Test_loss:0.298
Epoch:10, Train_acc:91.0%, Train_loss:0.220, Test_acc:89.5%, Test_loss:0.307
Epoch:11, Train_acc:90.8%, Train_loss:0.222, Test_acc:88.3%, Test_loss:0.316
Epoch:12, Train_acc:91.4%, Train_loss:0.210, Test_acc:83.3%, Test_loss:0.516
Epoch:13, Train_acc:91.5%, Train_loss:0.208, Test_acc:91.3%, Test_loss:0.247
Epoch:14, Train_acc:91.5%, Train_loss:0.206, Test_acc:90.1%, Test_loss:0.269
Epoch:15, Train_acc:92.0%, Train_loss:0.199, Test_acc:91.1%, Test_loss:0.242
Epoch:16, Train_acc:92.1%, Train_loss:0.194, Test_acc:89.4%, Test_loss:0.285
Epoch:17, Train_acc:92.4%, Train_loss:0.193, Test_acc:91.0%, Test_loss:0.229
Epoch:18, Train_acc:92.4%, Train_loss:0.188, Test_acc:88.0%, Test_loss:0.317
Epoch:19, Train_acc:92.7%, Train_loss:0.182, Test_acc:89.2%, Test_loss:0.285
Epoch:20, Train_acc:92.6%, Train_loss:0.182, Test_acc:78.5%, Test_loss:0.728
Done

5、結果可視化

import matplotlib.pyplot as plt
#隱藏警告
import warnings
warnings.filterwarnings("ignore")               #忽略警告信息epochs_range = range(epoches)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 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= Loss')
plt.show()

?
在這里插入圖片描述

?

在20輪測試集準確率變化比較大，從跑的幾次實驗來看，這次是偶然事件，測試集損失率后面一直穩定在0.3附件，測試準確率一直在0.8、0.89、0.90附件徘徊

6、模型評估

# 將參數加載到model當中 
best_model.load_state_dict(torch.load(PATH, map_location=device)) 
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)print(epoch_test_acc, epoch_test_loss)

0.9134651249533756 0.24670581874393283

本文來自互聯網用戶投稿，該文觀點僅代表作者本人，不代表本站立場。本站僅提供信息存儲空間服務，不擁有所有權，不承擔相關法律責任。
如若轉載，請注明出處：http://www.pswp.cn/news/898400.shtml
繁體地址，請注明出處：http://hk.pswp.cn/news/898400.shtml
英文地址，請注明出處：http://en.pswp.cn/news/898400.shtml

如若內容造成侵權/違法違規/事實不符，請聯系多彩編程網進行投訴反饋email:809451989@qq.com，一經查實，立即刪除！

相關文章

面試八股 —— Redis篇

面試八股 —— Redis篇

重點：緩存 和 分布式鎖 緩存（穿透，擊穿，雪崩） 降級可作為系統的保底策略，適用于穿透，擊穿，雪崩 1.緩存穿透 2.緩存擊穿 3.緩存雪崩 緩存——雙寫一致性 1.強一致性業務&#xff08…
閱讀更多...
mapbox-gl源碼中解析style加載地圖過程詳解

mapbox-gl源碼中解析style加載地圖過程詳解

我將結合 Mapbox GL JS 的源碼示例，一步一步講解 style 的解析和地圖加載過程，幫助大家深入理解其內部機制。 Mapbox GL JS 是一個強大的 Web 地圖庫，利用 WebGL 技術渲染交互式地圖。其核心功能之一是通過樣式（style&#xff09…
閱讀更多...
瑞薩RA系列使用JLink RTT Viewer輸出調試信息

瑞薩RA系列使用JLink RTT Viewer輸出調試信息

引言 還在用UART調試程序么？試試JLINK的RTT Viewer吧！不需占用UART端口、低資源暫用、實時性高延時微秒級，這么好的工具還有什么理由不用了！ 目錄 一、JLink RTT Viewer 簡介 二、軟件安裝 三、工程應用 3.1 SEGGER_RTT驅動包 3.2 手搓宏定義APP_PRINT 3.3 使用APP_…
閱讀更多...
MySQL 入門大全：查詢語言分類

MySQL 入門大全：查詢語言分類

🧑 博主簡介：CSDN博客專家，歷代文學網（PC端可以訪問：https://literature.sinhy.com/#/literature?__c1000，移動端可微信小程序搜索“歷代文學”）總架構師，15年工作經驗，…
閱讀更多...
1.Windows+vscode+cline+MCP配置

1.Windows+vscode+cline+MCP配置

文章目錄 1.簡介與資源2.在windows中安裝vscode及Cline插件1. 安裝vscode2. 安裝Cline插件3. 配置大語言模型3. 配置MCP步驟(windows) 1.簡介與資源 MCP官方開源倉庫 MCP合集網站 參考視頻 2.在windows中安裝vscode及Cline插件 1. 安裝vscode 2. 安裝Cline插件 Cline插件…
閱讀更多...
性能測試過程實時監控分析

性能測試過程實時監控分析

性能監控 前言一、查看性能測試結果的3大方式1、GUI界面報告插件2、命令行運行 html報告3、后端監聽器接入儀表盤 二、influxDB grafana jmeter測試監控大屏1、原理：2、linux環境中influxDB 安裝和配置3、jmerer后端監聽器連接influxDB4、linux環境總grafana環境搭…
閱讀更多...
【Linux我做主】淺談Shell及其原理

【Linux我做主】淺談Shell及其原理

淺談Linux中的Shell及其原理 Linux中Shell的運行原理github地址前言一、Linux內核與Shell的關系1.1 操作系統核心1.2 用戶與內核的隔離 二、Shell的演進與核心機制2.1 發展歷程2.2 核心功能解析2.3 shell的工作流程1. 用戶輸入命令2. 解析器拆分指令3. 擴展器處理動態內容變量替…
閱讀更多...
可視化圖解算法：鏈表中倒數（最后）k個結點

可視化圖解算法：鏈表中倒數（最后）k個結點

1. 題目 描述 輸入一個長度為 n 的鏈表，設鏈表中的元素的值為ai ，返回該鏈表中倒數第k個節點。 如果該鏈表長度小于k，請返回一個長度為 0 的鏈表。 數據范圍：0≤n≤105，0 ≤ai≤109，0 ≤k≤109 要求&am…
閱讀更多...
在線教育網站項目第四步：deepseek騙我， WSL2不能創建兩個獨立的Ubuntu，但我們能實現實例互訪及外部訪問

在線教育網站項目第四步：deepseek騙我， WSL2不能創建兩個獨立的Ubuntu，但我們能實現實例互訪及外部訪問

一、說明 上一章折騰了半天，搞出不少問題，今天我們在deepseek的幫助下，完成多個獨立ubuntu24.04實例的安裝，并完成固定ip，實踐證明，deepseek不靠譜，浪費我2個小時時間，我們將在下面實…
閱讀更多...
CMake 保姆級教程

CMake 保姆級教程

CMake 是一個跨平臺的構建工具，用于生成適合不同平臺和編譯器的構建系統文件（如 Makefile 或 Visual Studio 項目文件）。 在 Windows 下使用 CMake 構建項目時，CMake 會根據 CMakeLists.txt 文件生成適合 Windows 的構建系統文件&…
閱讀更多...
zabbix數據庫溯源

zabbix數據庫溯源

0x00 背景 zabbix數據庫如果密碼泄露被登錄并新增管理員如何快速發現？并進行溯源？ 本文介紹數據庫本身未開啟access log的情況。 0x01 實踐 Mysql 數據庫查insert SELECT * FROM sys.host_summary_by_statement_type where statement like %insert% 查…
閱讀更多...
Spring Boot集成PageHelper：輕松實現數據庫分頁功能

Spring Boot集成PageHelper：輕松實現數據庫分頁功能

Spring Boot集成PageHelper：輕松實現數據庫分頁功能 1. 為什么需要分頁？ 分頁是處理大數據量查詢的核心技術，其重要性體現在： 性能優化：避免單次查詢返回過多數據導致內存溢出或響應延遲。用戶體驗：前端展…
閱讀更多...
Spring Cloud之負載均衡之LoadBalance

Spring Cloud之負載均衡之LoadBalance

目錄 負載均衡 問題 步驟 現象 什么是負載均衡？ 負載均衡的一些實現 服務端負載均衡 客戶端負載均衡 使用Spring Cloud LoadBalance實現負載均衡 負載均衡策略 ?編輯 ?編輯LoadBalancer原理 服務部署 準備環境和數據 服務構建打包 啟動服務 上傳J…
閱讀更多...
數據無憂：自動備份策略全解析

數據無憂：自動備份策略全解析

引言 在信息化飛速發展的今天，數據已成為個人、企業乃至國家最為寶貴的資產之一。無論是日常辦公文檔、科研數據、客戶資料，還是個人隱私信息，一旦丟失或損壞，都可能帶來不可估量的損失。因此，備份文件作為數據安全的…
閱讀更多...
Latex2024安裝教程（附安裝包）Latex2024詳細圖文安裝教程

Latex2024安裝教程（附安裝包）Latex2024詳細圖文安裝教程

文章目錄 前言一、Latex2024下載二、Texlive 2024安裝教程1.準備安裝文件2.啟動安裝程序3.配置安裝選項4.開始安裝5.安裝完成6.TeX Live 2024 安裝后確認 三、Texstudio 安裝教程1.準備 Texstudio 安裝2.啟動 Texstudio 安裝向導3.選擇安裝位置4.等待安裝完成5.啟動 Texstudio6…
閱讀更多...
C++ 語法之函數和函數指針

C++ 語法之函數和函數指針

在上一章中 C 語法之 指針的一些應用說明-CSDN博客 我們了解了指針變量&#xff0c;int *p;取變量a的地址這些。 那么函數同樣也有個地址&#xff0c;直接輸出函數名就可以得到地址&#xff0c;如下&#xff1a; #include<iostream> using namespace std; void fun() …
閱讀更多...
centos【rockylinux】安裝【supervisor】的注意事項【完整版】

centos【rockylinux】安裝【supervisor】的注意事項【完整版】

重新加載 systemd 配置推薦使用pip的方式安裝 pip install supervisor 第二步&#xff1a;添加supervisord.conf配置文件 [unix_http_server] file/tmp/supervisor.sock ; UNIX socket 文件&#xff0c;supervisorctl 會使用 ;chmod0700 ; socket 文件的…
閱讀更多...
Spring Cloud Gateway 使用ribbon以及nacos實現灰度發布

Spring Cloud Gateway 使用ribbon以及nacos實現灰度發布

1、Spring Cloud Gateway配置文件 gateway:userId-limit: 1000 agent-bff:ribbon:NFLoadBalancerRuleClassName: com.anlitech.gateway.gray.GrayRule operator-bff:ribbon:NFLoadBalancerRuleClassName: com.anlitech.gateway.gray.GrayRule spring:cloud:gateway:locator:en…
閱讀更多...
關于“碰一碰發視頻”系統的技術開發文檔框架

關于“碰一碰發視頻”系統的技術開發文檔框架

以下是關于“碰一碰發視頻”系統的技術開發文檔框架&#xff0c;涵蓋核心功能、技術選型、開發流程和關鍵模塊設計&#xff0c;幫助您快速搭建一站式解決方案 --- 隨著短視頻平臺的興起&#xff0c;用戶的創作與分享需求日益增長。而如何讓視頻分享更加便捷、有趣&#xff0c…
閱讀更多...
基于django+vue的購物商城系統

基于django+vue的購物商城系統

開發語言&#xff1a;Python框架&#xff1a;djangoPython版本&#xff1a;python3.8數據庫&#xff1a;mysql 5.7數據庫工具&#xff1a;Navicat11開發軟件&#xff1a;PyCharm 系統展示 系統首頁 熱賣商品 優惠資訊 個人中心 后臺登錄 管理員功能界面 用戶管理 商品分類管理…
閱讀更多...
最新文章

Copyright @ 2022~2023