<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Fan2g]]></title><description><![CDATA[不积跬步，无以至千里]]></description><link>https://blog.youkuaiyun.com/u014135752</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; u014135752]]></copyright><item><title><![CDATA[特征工程——连续特征与离散特征处理方法]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80789251</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80789251</guid><author>u014135752</author><pubDate>Sun, 24 Jun 2018 10:14:06 +0800</pubDate><description><![CDATA[特征的标准化（归一化）

z-score标准化：这是最常见的特征预处理方式，基本所有的线性模型在拟合的时候都会做 z-score标准化。具体的方法是求出样本特征x的均值mean和标准差std，然后用（x-mean)/std来代替原特征。这样特征就变成了均值为0，方差为1了。在sklearn中，我们可以用StandardScaler来做z-score标准化。当然，如果我们是用pandas做数据预处理...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-98——Validate Binary Search Tree（验证搜索二叉树）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80726786</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80726786</guid><author>u014135752</author><pubDate>Mon, 18 Jun 2018 21:34:47 +0800</pubDate><description><![CDATA[链接：点击打开链接题目：给定一个二叉树，判断其是否是一个有效的二叉搜索树。一个二叉搜索树具有如下特征：节点的左子树只包含小于当前节点的数。节点的右子树只包含大于当前节点的数。所有左子树和右子树自身必须也是二叉搜索树。Example:示例1输入:
    2
   / \
  1   3
输出: true示例2    5
   / \
  1   4
     / \
    3   6
Outpu...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-——Maximum Depth of Binary Tree（二叉树的最大深度）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80726325</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80726325</guid><author>u014135752</author><pubDate>Mon, 18 Jun 2018 20:44:22 +0800</pubDate><description><![CDATA[链接：点击打开链接题目：给定一个二叉树，找出其最大深度。二叉树的深度为根节点到最远叶子节点的最长路径上的节点数。Example:给定二叉树 [3,9,20,null,null,15,7]，    3
   / \
  9  20
    /  \
   15   7返回它的最大深度 3 。解析:如果根节点为空，则深度为0，返回0，递归的出口如果根节点不为空，那么深度至少为1，然后我们求他们左右子树...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-141——Linked List Cycle（判断链表中是否有环）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80713668</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80713668</guid><author>u014135752</author><pubDate>Sat, 16 Jun 2018 16:47:44 +0800</pubDate><description><![CDATA[链接：点击打开链接题目：Given a linked list, determine if it has a cycle in it.Notes:Can you solve it without using extra space?解析:由于每一个父亲只有可能有一个孩子，故这里的环实际上是指list中某一个节点的孩子同时也是它自己或者他的祖先。 这个问题需要注意几种情况：1. 空链表不成环2. 一...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-234——Palindrome Linked List（回文链表）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80713276</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80713276</guid><author>u014135752</author><pubDate>Sat, 16 Jun 2018 15:49:48 +0800</pubDate><description><![CDATA[链接：点击打开链接题目：Given a singly linked list, determine if it is a palindrome.Example:Example 1:Input: 1-&amp;gt;2
Output: falseExample 2:Input: 1-&amp;gt;2-&amp;gt;2-&amp;gt;1
Output: trueNotes:Could you do it in O(n) tim...]]></description><category></category></item><item><title><![CDATA[利用CNN对MNIST数据集识别(keras框架)]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80700975</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80700975</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:32:14 +0800</pubDate><description><![CDATA[1、数据集下载https://blog.youkuaiyun.com/wuzhichenggo/article/details/793321282、代码执行from tensorflow.examples.tutorials.mnist import input_data

#导入数据
dir = r'F:\dataset\mnist'
mnist = input_data.read_data_sets(di...]]></description><category></category></item><item><title><![CDATA[LeNet、AlexNet、GoogLeNet、VGG、ResNet]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/79975967</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/79975967</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:23:26 +0800</pubDate><description><![CDATA[https://zhuanlan.zhihu.com/p/22094600 
https://blog.youkuaiyun.com/u013989576/article/details/71600795 
https://blog.youkuaiyun.com/wcy12341189/article/details/56281618 
https://blog.youkuaiyun.com/sunbaigui/article/de...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-206——Reverse Linked List（反转链表）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80698516</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80698516</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:19:38 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/explore/interview/card/top-interview-questions-easy/93/linked-list/560/

题目：

Reverse a singly linked list.

Example:

Input: 1-&amp;amp;amp;gt;2-&amp;amp;amp;gt;3-&amp;amp;amp;gt;4-&amp;amp;amp;gt;5-&amp;amp;amp;gt;NULL
Output: 5]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-21——Merge Two Sorted Lists（合并连个有序链表）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80698424</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80698424</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:19:23 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/merge-two-sorted-lists/description/

题目：

Merge two sorted linked lists and return it as a new list. The new list should be made by splicing together the nodes of th...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-28——Implement strStr()]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80672123</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80672123</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:19:12 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/implement-strstr/description/

题目：

Implement strStr(). 
Return the index of the first occurrence of needle in haystack, or -1 if needle is not part of haystack.

Ex...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-19——Remove Nth Node From End of List（ 删除链表的倒数第N个节点 ）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80686244</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80686244</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:18:58 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/remove-nth-node-from-end-of-list/description/

题目：

给定一个链表，删除链表的倒数第 n 个节点，并且返回链表的头结点。

Example:

给定一个链表: 1-&amp;amp;gt;2-&amp;amp;gt;3-&amp;amp;gt;4-&amp;amp;gt;5, 和 n = 2.

当删除了倒数第二个节点后，链表变为 1-&amp;amp;gt...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-26—— Remove Duplicates from Sorted Array（删除排序数组中的重复项）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80671955</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80671955</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:18:44 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/remove-duplicates-from-sorted-array/description/

题目：

Given a sorted array nums, remove the duplicates in-place such that each element appear only once and return t...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-1——Two sum(两数之和)]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80671684</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80671684</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:18:31 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/two-sum/description/

题目：

Given an array of integers, return indices of the two numbers such that they add up to a specific target. 
You may assume that each input ...]]></description><category></category></item><item><title><![CDATA[LeetCode刷题-7—— Reverse Integer(反转整数)]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80671766</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80671766</guid><author>u014135752</author><pubDate>Fri, 15 Jun 2018 09:18:18 +0800</pubDate><description><![CDATA[链接：

https://leetcode.com/problems/reverse-integer/description/

题目：

Given a 32-bit signed integer, reverse digits of an integer.

Example:

Example 1:



Input: 123
Output: 321

Example 2:



Input:...]]></description><category></category></item><item><title><![CDATA[梯度提升决策树——Gradient Boosting Decision Tree（GBDT）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80606239</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80606239</guid><author>u014135752</author><pubDate>Fri, 08 Jun 2018 11:16:15 +0800</pubDate><description><![CDATA[https://www.cnblogs.com/pinard/p/6140514.html#!comments#undefined 
https://blog.youkuaiyun.com/shine19930820/article/details/65633436 
https://blog.youkuaiyun.com/shine19930820/article/details/62233166#%E6%9C%80%...]]></description><category></category></item><item><title><![CDATA[随机森林算法——Random Forest（RF）]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80612184</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80612184</guid><author>u014135752</author><pubDate>Thu, 07 Jun 2018 17:09:11 +0800</pubDate><description><![CDATA[Bagging和Boosting 概念及区别

随机森林属于集成学习（Ensemble Learning）中的bagging算法。在集成学习中，主要分为bagging算法和boosting算法。我们先看看这两种方法的特点和区别。



Bagging

bagging的算法过程如下：


从原始样本集中使用Bootstraping方法随机抽取n个训练样本，共进行k轮抽取，得到k个训练集。（k个训练...]]></description><category></category></item><item><title><![CDATA[基于Keras2《面向小数据集构建图像分类模型》——Kaggle猫狗数据集]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80072493</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80072493</guid><author>u014135752</author><pubDate>Tue, 24 Apr 2018 23:20:31 +0800</pubDate><description><![CDATA[概述

在本文中，将使用VGG-16模型提供一种面向小数据集（几百张到几千张图片）构造高效、实用的图像分类器的方法并给出试验结果。

本文将探讨如下几种方法：


从图片中直接训练一个小网络（作为基准方法）
利用预训练网络的bottleneck（瓶颈）特征
fine-tune预训练网络的高层


本文需要使用的Keras模块有：


fit_generator：用于从Python生成器中训练网络
...]]></description><category></category></item><item><title><![CDATA[使用PyCharm进行远程开发和调试]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/80029979</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/80029979</guid><author>u014135752</author><pubDate>Sat, 21 Apr 2018 15:14:56 +0800</pubDate><description><![CDATA[转载自：https://blog.youkuaiyun.com/ll641058431/article/details/53049453


                            


使用PyCharm进行远程开发和调试



你是否经常要在Windows 7或MAC OS X上面开发Python或Web应用程序，但是它们最后需要在linux上面来运行呢？ 我们经常会碰到开发时没有问题但是...]]></description><category></category></item><item><title><![CDATA[AlexNet]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/79978097</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/79978097</guid><author>u014135752</author><pubDate>Tue, 17 Apr 2018 17:36:13 +0800</pubDate><description><![CDATA[AlexNet模型解读

 
AlexNet共有八层，有60M以上的参数量。 
前五层为卷积层：convolutional layer； 
后三层为全连接层，fully connected layer。 
最后一个全连接层输出具有1000个输出的softmax。（这个不懂什么意思的参考机器学习理论中的softmax regression，就是一个多类的logistic regression）

大...]]></description><category></category></item><item><title><![CDATA[python错误——SyntaxError: EOL while scanning string literal]]></title><link>https://blog.youkuaiyun.com/u014135752/article/details/79966412</link><guid>https://blog.youkuaiyun.com/u014135752/article/details/79966412</guid><author>u014135752</author><pubDate>Mon, 16 Apr 2018 20:57:26 +0800</pubDate><description><![CDATA[背景

python 2.7.2 中想要通过urllib下载百度空间中的图片，存到本地。

参考这里：

http://www.cnblogs.com/huangcong/archive/2011/09/03/2165565.html

如此实现：

saved_pic = r”E:\WebServer\WordPress\to_wp\hi-baidu-mover_v2\pic\” + picNa...]]></description><category></category></item></channel></rss>