<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Lin_drink_caffe的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Lin_drink_caffe</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Lin_drink_caffe]]></copyright><item><title><![CDATA[GAN的简单应用]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/74639966</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/74639966</guid><author>Lin_drink_caffe</author><pubDate>Fri, 07 Jul 2017 11:09:47 +0800</pubDate><description><![CDATA[GAN学习指南：从原理入门到制作生成Demo，总共分几步？ 
https://www.leiphone.com/news/201701/yZvIqK8VbxoYejLl.html?viewType=weixin]]></description><category></category></item><item><title><![CDATA[机器学习： batch normalization]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/74637844</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/74637844</guid><author>Lin_drink_caffe</author><pubDate>Fri, 07 Jul 2017 10:49:43 +0800</pubDate><description><![CDATA[参考文章： 
1，说明http://blog.youkuaiyun.com/elaine_bao/article/details/50890491 
2，代码http://blog.youkuaiyun.com/elaine_bao/article/details/50923198 
3，原文paper：https://arxiv.org/pdf/1502.03167.pdf]]></description><category></category></item><item><title><![CDATA[C++ 希尔排序]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73958039</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73958039</guid><author>Lin_drink_caffe</author><pubDate>Fri, 30 Jun 2017 11:12:43 +0800</pubDate><description><![CDATA[ShellSort.h文件#ifndef SHELLSORT_H
#define SHELLSORT_H
#include <vector>
#include <iostream>
using std::vector;
using std::cout;
using std::endl;template <typename T>
class ShellSort
{
private:
    unsig]]></description><category></category></item><item><title><![CDATA[直接插入排序 C++]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73862546</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73862546</guid><author>Lin_drink_caffe</author><pubDate>Wed, 28 Jun 2017 18:24:49 +0800</pubDate><description><![CDATA[//直接插入排序
#include <iostream>
using namespace std;void insertSort(int arr[], int length)
{
    int i,j,key;
    for(i=0;i<length;i++){
        key = arr[i];
        for(j=i-1;j>=0;j--){            if(ke]]></description><category></category></item><item><title><![CDATA[简单选择排序 C++]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73838617</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73838617</guid><author>Lin_drink_caffe</author><pubDate>Wed, 28 Jun 2017 13:28:01 +0800</pubDate><description><![CDATA[//简单选择排序
#include <iostream>using namespace std;void selectSort(int r[], int n){    int i,min,j,temp;
    //第i遍扫描
    for(i=1;i<n;i++){
        min=i;
        for(j=i+1;j<n;j++)
        {
            i]]></description><category></category></item><item><title><![CDATA[C++ 冒泡排序]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73810665</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73810665</guid><author>Lin_drink_caffe</author><pubDate>Tue, 27 Jun 2017 17:39:23 +0800</pubDate><description><![CDATA[#include <iostream>
using namespace std;void my_print(int* pData,int count){
    for(int i=0;i<count;i++){
        cout<<pData[i]<<" ";
    }
    cout<<endl;
}void BubbleSort(int* pData,int count)
{]]></description><category></category></item><item><title><![CDATA[计算机视觉、机器学习、模式识别、图像处理领域国内外期刊、会议汇总]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73773944</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73773944</guid><author>Lin_drink_caffe</author><pubDate>Tue, 27 Jun 2017 11:08:00 +0800</pubDate><description><![CDATA[一、重要国际期刊
IEEE Transactions on Pattern Analysis and Machine Intelligence   (PAMI)    (IEEE模式分析与机器智能汇刊) 
http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=34
IEEE Transactions on ImageProcessing(IP]]></description><category></category></item><item><title><![CDATA[评价指标]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73691955</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73691955</guid><author>Lin_drink_caffe</author><pubDate>Sat, 24 Jun 2017 14:53:16 +0800</pubDate><description><![CDATA[机器学习常用的评价指标]]></description><category></category></item><item><title><![CDATA[最小生成树 prim算法]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73611126</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73611126</guid><author>Lin_drink_caffe</author><pubDate>Thu, 22 Jun 2017 21:10:46 +0800</pubDate><description><![CDATA[一、prim算法核心问题：用最小的成本连接图中所有的顶点 
参考文献： 
1、《大话数据结构》 
2、http://blog.youkuaiyun.com/yeruby/article/details/38615045#
//prim算法
#include <iostream>
#include <fstream>
using namespace std;#define MAX 100
#define MAXC]]></description><category></category></item><item><title><![CDATA[机器学习： 激活函数的认识]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73350356</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73350356</guid><author>Lin_drink_caffe</author><pubDate>Fri, 16 Jun 2017 20:55:20 +0800</pubDate><description><![CDATA[一、必要性激活函数的主要作用是提供网络的非线性建模能力。如果没有激活函数，那么该网络仅能够表达线性映射，此时即便有再多的隐藏层，其整个网络跟单层神经网络也是等价的。因此也可以认为，只有加入了激活函数之后，深度神经网络才具备了分层的非线性映射学习能力。二、种类
sigmoid函数 
 
特别注意sigmoid函数的导数求法： 
σ′(x)=σ(x)∗(1−σ(x))\sigma^{'}(x)=\sig]]></description><category></category></item><item><title><![CDATA[机器学习：python Virtual environment 流程]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73180026</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73180026</guid><author>Lin_drink_caffe</author><pubDate>Tue, 13 Jun 2017 15:00:42 +0800</pubDate><description><![CDATA[具体描述来自 斯坦福大学cs213n课程cd assignment1
sudo pip install virtualenv      # This may already be installed
virtualenv -p python3 .env       # Create a virtual environment (python3)
# Note: you can also use "v]]></description><category></category></item><item><title><![CDATA[计算机视觉资料]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73065615</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73065615</guid><author>Lin_drink_caffe</author><pubDate>Sun, 11 Jun 2017 21:42:10 +0800</pubDate><description><![CDATA[斯坦福大学 cs231n课件以及作业
http://vision.stanford.edu/teaching/cs231n/syllabus.html]]></description><category></category></item><item><title><![CDATA[机器学习：从二项logistic distribution 到 softmax loss function]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73008230</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/73008230</guid><author>Lin_drink_caffe</author><pubDate>Sat, 10 Jun 2017 21:20:28 +0800</pubDate><description><![CDATA[一、binomial logistic distribution 
xx是输入的特征向量值，ww是权值，w∗xw*x是向量的内积，Y∈{0,1}Y \in \{0,1\}是输出，利用上式求出两个概率值，将xx划分为值大的那一类。
对数几率（log odds） 
一个事件的几率（odds）是指该事件发生的概率pp 与不发生的概率（1−p1-p）比值，即p1−p\frac{p}{1-p}。那么对数几率为]]></description><category></category></item><item><title><![CDATA[一元线性回归，多元线性回归，MATLAB]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72970951</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72970951</guid><author>Lin_drink_caffe</author><pubDate>Fri, 09 Jun 2017 20:43:44 +0800</pubDate><description><![CDATA[一元线性回归clear;clc;
%一元线性回归
x=[143,145,146,147,149,150,153,154,155,156,157,158,159,160,162,164];
y=[88,85,88,91,92,93,93,95,96,98,97,96,98,99,100,102];
X=[ones(length(y),1),x'];
Y=y';
%b 参数
%bint 参数对于的置信区]]></description><category></category></item><item><title><![CDATA[机器学习知识总结：代价函数与经验风险、结构风险最小化]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72861356</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72861356</guid><author>Lin_drink_caffe</author><pubDate>Sun, 04 Jun 2017 20:56:11 +0800</pubDate><description><![CDATA[一、代价函数代价函数（cost function）用来度量预测错误，损失函数越小模型越好，常见的有4种：
0-1损失函数： 平方损失函数： 绝对损失函数： 对数损失函数： 二、经验风险最小化与结构风险最小化
经验风险最小化（empirical risk minimization，EMR） 
当样本容量很大的时候，经验风险最小化的策略认为，经验风险最小的模型是最优的模型： 
 
当样本容量很小的时候，]]></description><category></category></item><item><title><![CDATA[Leetcode:Roman to Integer]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72852232</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72852232</guid><author>Lin_drink_caffe</author><pubDate>Sat, 03 Jun 2017 15:15:47 +0800</pubDate><description><![CDATA[/*
 * Given a roman numeral, convert it to an integer. * Input is guaranteed to be within the range from 1 to 3999.
 */
#include <iostream>
using namespace std;class Solution {  
public:  
    int roma]]></description><category></category></item><item><title><![CDATA[电脑数据存放]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72784349</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72784349</guid><author>Lin_drink_caffe</author><pubDate>Sat, 27 May 2017 20:59:26 +0800</pubDate><description><![CDATA[1、ROM芯片它是BIOS（基本输入输出系统），固化于计算机内主板上，保存着计算机最重要的基本输入输出的程序、开机后自检程序和系统自启动程序，它可从CMOS中读写系统设置的具体信息。 其主要功能是为计算机提供最底层的、最直接的硬件设置和控制，平时靠纽扣电池供电维持CMOS数据记忆，没电或损坏时恢复到出厂设置；2、内存条（RAM）是CPU可通过总线寻址，并进行读写操作的电脑部件，写入的数据将在断电后彻]]></description><category></category></item><item><title><![CDATA[leetcode:Maximum Subarray]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72633479</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72633479</guid><author>Lin_drink_caffe</author><pubDate>Tue, 23 May 2017 00:21:55 +0800</pubDate><description><![CDATA[#include <iostream>
#include <algorithm>
#include <vector>
using namespace std;class Solution{
public:
    int maxSubArray(vector<int>& nums){
        //int n = nums.size();
        if(nums.empty()||!n]]></description><category></category></item><item><title><![CDATA[Leetcode:Climbing Stairs]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72631004</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72631004</guid><author>Lin_drink_caffe</author><pubDate>Mon, 22 May 2017 22:31:47 +0800</pubDate><description><![CDATA[#include <iostream>
#include <vector>
using namespace std;class Solution1
{
public:
    int climbStairs(int n){
        if(n<=2)
        {
            return n;
        }
        else
        {]]></description><category></category></item><item><title><![CDATA[leetcode: Search Insert Position]]></title><link>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72512147</link><guid>https://blog.youkuaiyun.com/Lin_drink_caffe/article/details/72512147</guid><author>Lin_drink_caffe</author><pubDate>Thu, 18 May 2017 23:22:22 +0800</pubDate><description><![CDATA[//https://leetcode.com/problems/search-insert-position/#/description
#include <iostream>
#include <vector>
using namespace std;
class Solution {
public:
    int searchInsert(vector<int>& nums, int targ]]></description><category></category></item></channel></rss>