<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[likesomething1的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/likesomething1</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; likesomething1]]></copyright><item><title><![CDATA[用OpenCV的边缘检测]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/126090145</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/126090145</guid><author>likesomething1</author><pubDate>Sun, 31 Jul 2022 20:39:27 +0800</pubDate><description><![CDATA[边缘检测]]></description><category></category></item><item><title><![CDATA[Linux使用CMake构建OpenCV、onnxruntime模型部署项目]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/125643598</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/125643598</guid><author>likesomething1</author><pubDate>Wed, 06 Jul 2022 17:32:04 +0800</pubDate><description><![CDATA[Linux使用CMake构建需要用到OpenCV、onnxruntime库的模型部署项目，可省去添加lib、include路径，编译时也不用添加动态连接库。]]></description><category></category></item><item><title><![CDATA[Linux 安装TensorRT、Pycuda]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/125619478</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/125619478</guid><author>likesomething1</author><pubDate>Tue, 05 Jul 2022 14:56:06 +0800</pubDate><description><![CDATA[记录安装TensorRt、pycuda的过程
]]></description><category></category></item><item><title><![CDATA[部署Pytorch模型到C++环境]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/125543214</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/125543214</guid><author>likesomething1</author><pubDate>Thu, 30 Jun 2022 18:26:47 +0800</pubDate><description><![CDATA[由于工作原因需要部署Pytorch模型到c++环境下，目前大概有三种方式。
1、pytorch转成onnx文件后，通过opencv读取。
2、pytroch转成onnx文件后，通过onnxruntime读取。
3、利用libtorch库，也就是pytorch的c++版。首先的将pytorch训练好的模型导出onnx文件。...............]]></description><category></category></item><item><title><![CDATA[压缩感知中贪婪算法之MP和OMP]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/116299495</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/116299495</guid><author>likesomething1</author><pubDate>Fri, 30 Apr 2021 12:20:47 +0800</pubDate><description><![CDATA[压缩感知中贪婪算法之MP和OMP
1、	Matching Pusuit(匹配追踪):

MP是最简单的追踪算法之一，每次只从A中选取一列，每步迭代时就更新该列所对应的系数。但是，MP一般会反复选择矩阵A的同一列，以进一步减小逼近的误差。因此，算法复杂度较高。
Matlab代码实现如下：
function x = CS_MP(y, A, t )
  % Matching Pursuit
  %
  %   y = Phi * x
  %   x = Psi * theta
  %	  y = Phi*Psi ]]></description><category></category></item><item><title><![CDATA[蚂蚁算法matlab实现TSP]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/108882414</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/108882414</guid><author>likesomething1</author><pubDate>Wed, 30 Sep 2020 11:23:17 +0800</pubDate><description><![CDATA[clear;
close all;
clc;
n = 35; %城市的数量
x = rand(1,n)*10;
y = rand(1,n)*10;
D = zeros(n,n);
for i = 1:n
for j = 1:n
if i~= j
D(i,j) = sqrt(power(x(i)-x(j),2) + power(y(i)-y(j),2));
else
D(i,j) = eps;
end
end
end
m = 80; %蚂蚁的数量
alpha = 2;
beta = 6;
rho = 0.2;]]></description><category></category></item><item><title><![CDATA[统计学习方法(第二版)笔记]]></title><link>https://blog.youkuaiyun.com/likesomething1/article/details/107059619</link><guid>https://blog.youkuaiyun.com/likesomething1/article/details/107059619</guid><author>likesomething1</author><pubDate>Wed, 01 Jul 2020 11:31:13 +0800</pubDate><description><![CDATA[一、【统计学习方法概论】
1.机器学习的定义，以及与数据挖掘的关系
2.统计学习的定义、特点和目的
3.监督学习的基本概念
3.1 输入空间、特征空间与输出空间
I.输入空间：将输入所有可能取值的集合
II.输出空间：将输出所有可能取值的集合
III.特征空间：所有特征向量存在的空间
3.2 联合概率分布
I.监督学习假设输入与输出的随机变量X，Y遵循联合概率分布P(X,Y)
II.训练和测试数据看作是P(X,Y)独立同分布产生的
3.3 假设空间
I.监督学习的目的在于学习一个由输入到输出的映射，这一映射]]></description><category></category></item></channel></rss>