<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[sunboon的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/sunboon</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; sunboon]]></copyright><item><title><![CDATA[yolov5实现天气识别]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/145346651</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/145346651</guid><author>sunboon</author><pubDate>Fri, 24 Jan 2025 22:20:07 +0800</pubDate><description><![CDATA[std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸。]]></description><category></category></item><item><title><![CDATA[YOLOv5-C3模块实现]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/145176194</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/145176194</guid><author>sunboon</author><pubDate>Thu, 16 Jan 2025 11:30:11 +0800</pubDate><description><![CDATA[std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸。]]></description><category></category></item><item><title><![CDATA[【无标题】]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/145041964</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/145041964</guid><author>sunboon</author><pubDate>Thu, 09 Jan 2025 21:02:53 +0800</pubDate><description><![CDATA[std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸。]]></description><category></category></item><item><title><![CDATA[【无标题】VGG-16算法-Pytorch实现人脸识别]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144893418</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144893418</guid><author>sunboon</author><pubDate>Thu, 02 Jan 2025 20:12:32 +0800</pubDate><description><![CDATA[model.classifier._modules['6'] = nn.Linear(4096,len(classeNames)) # 修改vgg16模型中最后一层全连接层，输出目标类别个数#这里边模型可以解冻几层，进一步提高准确率。std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。]]></description><category></category></item><item><title><![CDATA[运动鞋识别]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144717884</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144717884</guid><author>sunboon</author><pubDate>Wed, 25 Dec 2024 13:55:21 +0800</pubDate><description><![CDATA[🍨 本文为🔗365天深度学习训练营中的学习记录博客
	🍖 原作者：K同学啊
	
	一、导入数据
	
	
	import torch
	import torch.nn as nn
	import torchvision.transforms as transforms
	from torchvision import datasets
	import torchvision.models as models
	from PIL import Image

	import os
	import pathlib
]]></description><category></category></item><item><title><![CDATA[pytorch猴痘病识别]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144592360</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144592360</guid><author>sunboon</author><pubDate>Thu, 19 Dec 2024 21:58:29 +0800</pubDate><description><![CDATA[本篇与之前不同的地方在于卷积层换成了weights = ResNet18_Weights.DEFAULT  # 使用默认的预训练权重model = models.resnet18(weights=weights)  # 加载预训练的ResNet18模型来提高测试准确率以及加入了    if epoch_test_acc > best_acc:best_model_wts = model.state_dict()  # 保存当前模型的权重筛选出最好的模型参数，并用该模型参数加载图片运行代码。]]></description><category></category></item><item><title><![CDATA[pytorch天气识别学习]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144402526</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144402526</guid><author>sunboon</author><pubDate>Wed, 11 Dec 2024 17:08:12 +0800</pubDate><description><![CDATA[与之前的不同点：导入图片方面采用了新的导入方式，这里我用的是os.path.join的拼接方式，可以避免不同系统拼接方式不一样。然后多次采用列表推导的方法写入图片路径，图片文件名的一一对应，最后加载出图片。然后经过归一化等一系列操作对图像进行处理并分为测试和训练集。]]></description><category></category></item><item><title><![CDATA[CIFAR10彩色图片识别]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144221586</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144221586</guid><author>sunboon</author><pubDate>Wed, 04 Dec 2024 16:55:34 +0800</pubDate><description><![CDATA[原代码在苹果系统里依旧会报错，不能直接下载，这里我依旧采用的是手动下载以后，通过代码导入数据。import os"""手动加载 CIFAR-10 数据集:param root: 数据集文件路径:param train: 是否加载训练集（True）还是测试集（False）:param transform: 可选的图像预处理方法"""# 训练数据文件列表else:# 加载数据和标签。]]></description><category></category></item><item><title><![CDATA[pytorch手写识别学习]]></title><link>https://blog.youkuaiyun.com/sunboon/article/details/144003589</link><guid>https://blog.youkuaiyun.com/sunboon/article/details/144003589</guid><author>sunboon</author><pubDate>Sun, 24 Nov 2024 11:27:55 +0800</pubDate><description><![CDATA[root='/Users/qiang/PycharmProjects/PythonProject3/data',                      # 替换为你的数据集路径download=False                    # 设置为 False，不再自动下载root='/Users/qiang/PycharmProjects/PythonProject3/data',                      # 替换为你的数据集路径。]]></description><category></category></item></channel></rss>