<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[baidu_38812770的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/baidu_38812770</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; baidu_38812770]]></copyright><item><title><![CDATA[数据分析实战之数据预处理]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103568890</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103568890</guid><author>baidu_38812770</author><pubDate>Mon, 16 Dec 2019 20:16:36 +0800</pubDate><description><![CDATA[1.餐饮销量数的统计量情况


import pandas as pd
import matplotlib.pyplot as plt
# 餐饮销量数的统计量情况
catering_sale = '../data/chapter3/demo/data/catering_sale.xls'  # 餐饮数据
data = pd.read_excel(catering_sale, index_col='...]]></description><category></category></item><item><title><![CDATA[深度学习]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103557144</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103557144</guid><author>baidu_38812770</author><pubDate>Mon, 16 Dec 2019 16:47:26 +0800</pubDate><description><![CDATA[1.神经网络实现分类


# 设计神经网络结构，按照已定结构训练神经网络实现分类业务
import numpy as np
import matplotlib.pyplot as mp


# sigmoid函数
def active(x):
    return 1/(1+np.exp(-x))


# 导函数
def backward(x):
    return x*(1-x)


# 向前...]]></description><category></category></item><item><title><![CDATA[机器学习6之图像识别]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103526418</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103526418</guid><author>baidu_38812770</author><pubDate>Fri, 13 Dec 2019 17:08:29 +0800</pubDate><description><![CDATA[1.OpenCv基础


import numpy as np
import cv2 as cv

# 读取图片
original = cv.imread('../data/ml_data/forest.jpg')
cv.imshow('Original', original)
# 显示图片某个颜色通道的图像
blue = np.zeros_like(original)
blue[:, :, 0]...]]></description><category></category></item><item><title><![CDATA[机器学习5之自然语言处理]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103503653</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103503653</guid><author>baidu_38812770</author><pubDate>Fri, 13 Dec 2019 11:23:03 +0800</pubDate><description><![CDATA[1.处理流程


	语音识别
	
	
	自然语言处理 - 语义分析
	
	
	逻辑分析 - 结合业务场景与上下文
	
	
	自然语言处理 - 分析结果生成自然语言文本
	
	
	语音合成
	
自然语言处理的常用处理过程：

先针对训练文本进行分词处理（词干提取、原型提取），统计词频，通过词频-逆文档频率算法获得该词对样本语义的贡献，根据每个词的贡献力度，构建有监督分类学习模型。把测试样本交给模型处...]]></description><category></category></item><item><title><![CDATA[机器学习4之聚类]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103497045</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103497045</guid><author>baidu_38812770</author><pubDate>Thu, 12 Dec 2019 09:51:58 +0800</pubDate><description><![CDATA[1.k-means算法

第一步：随机选择k个样本作为k个聚类的中心，计算每个样本到各个聚类中心的欧氏距离，将该样本分配到与之距离最近的聚类中心所在的类别中。

第二步：根据第一步所得到的聚类划分，分别计算每个聚类的几何中心，将几何中心作为新的聚类中心，重复第一步，直到计算所得几何中心与聚类中心重合或接近重合为止。


# 加载multiple3.txt，基于K均值算法完成样本的聚类

impor...]]></description><category></category></item><item><title><![CDATA[机器学习3之分类]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103486295</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103486295</guid><author>baidu_38812770</author><pubDate>Wed, 11 Dec 2019 17:53:52 +0800</pubDate><description><![CDATA[1.人工分类


import numpy as np
import matplotlib.pyplot as mp
x = np.array([
    [3, 1],
    [2, 5],
    [1, 8],
    [6, 4],
    [5, 2],
    [3, 5],
    [4, 7],
    [4, -1]])
y = np.array([0, 1, 1, 0, 0,...]]></description><category></category></item><item><title><![CDATA[机器学习2之回归]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103481663</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103481663</guid><author>baidu_38812770</author><pubDate>Tue, 10 Dec 2019 20:33:29 +0800</pubDate><description><![CDATA[1.线性回归


# 基于线性回归训练single.txt中的训练样本，使用模型预测测试样本。
import numpy as np
import sklearn.linear_model as lm
import matplotlib.pyplot as mp
# 采集数据
x, y = np.loadtxt('../data/ml_data/single.txt',
             ...]]></description><category></category></item><item><title><![CDATA[机器学习1之基础]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103455380</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103455380</guid><author>baidu_38812770</author><pubDate>Mon, 09 Dec 2019 17:51:22 +0800</pubDate><description><![CDATA[1.典型应用

 股价预测、推荐引擎、自然语言识别、语音识别、图像识别、人脸识别

2.机器学习的基本问题

1)回归问题：根据已知的输入和输出寻找某种性能最佳的模型，将未知输出的输入代入模型，得到连续的输出。

2)分类问题：根据已知的输入和输出寻找某种性能最佳的模型，将未知输出的输入代入模型，得到离散的输出。

3)聚类问题：根据已知输入的相似程度，将其划分为不同的群落。

3.基本步骤...]]></description><category></category></item><item><title><![CDATA[数据分析8之pandas]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103398444</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103398444</guid><author>baidu_38812770</author><pubDate>Thu, 05 Dec 2019 17:46:20 +0800</pubDate><description><![CDATA[1.pandas核心数据结构


"""
Series可以理解为一个一维的数组，只是index名称可以自己改动。
类似于定长的有序字典，有Index和 value
"""
import numpy as np
import pandas as pd

data = np.array(['a', 'b', 'c', 'd'])
s = pd.Series(data)
print(s)

s = pd...]]></description><category></category></item><item><title><![CDATA[数据分析7之数学基础2]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103381101</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103381101</guid><author>baidu_38812770</author><pubDate>Thu, 05 Dec 2019 08:52:46 +0800</pubDate><description><![CDATA[1.奇异值分解


"""
有一个矩阵M，可以分解为3个矩阵U、S、V，使得U x S x V
等于M。U与V都是正交矩阵（乘以自身的转置矩阵结果为单位矩
阵）。那么S矩阵主对角线上的元素称为矩阵M的奇异值，其它元素均为0。
"""
import numpy as np

M = np.mat('4 11 14; 8 7 -2')
U, sv, V = np.linalg.svd(M, full_...]]></description><category></category></item><item><title><![CDATA[数据分析6之数学基础]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103361709</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103361709</guid><author>baidu_38812770</author><pubDate>Tue, 03 Dec 2019 17:51:56 +0800</pubDate><description><![CDATA[1.符号数组


import matplotlib.pyplot as mp
import numpy as np
import matplotlib.dates as md
import datetime as dt

# 若相比上一天的收盘价上涨，则为正成交量；若相比上一天的收盘价下跌，则为负成交量。
# 绘制OBV柱状图

# 日期转换函数
def dmy2ymd(dmy):
    dm...]]></description><category></category></item><item><title><![CDATA[数据分析5之线性模型]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103345286</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103345286</guid><author>baidu_38812770</author><pubDate>Mon, 02 Dec 2019 17:30:28 +0800</pubDate><description><![CDATA[1，线性预测


import numpy as np
import matplotlib.pyplot as plt
import datetime as dt
import matplotlib.dates as md


# 日期转换函数
def dmy2ymd(dmy):
    dmy = str(dmy, encoding='utf-8')
    time = dt.datetime...]]></description><category></category></item><item><title><![CDATA[数据分析4之numpy常用函数]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103298790</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103298790</guid><author>baidu_38812770</author><pubDate>Fri, 29 Nov 2019 17:42:56 +0800</pubDate><description><![CDATA[1.加载文件


import numpy as np
import matplotlib.pyplot as plt
import datetime as dt
import matplotlib.dates as md


# 日期转换函数
# def dmy2ymd(dmy):
#     dmy = str(dmy, encoding='utf-8')
#     time = dt.da...]]></description><category></category></item><item><title><![CDATA[数据分析3之几种常用图]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103287634</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103287634</guid><author>baidu_38812770</author><pubDate>Thu, 28 Nov 2019 16:46:59 +0800</pubDate><description><![CDATA[1.填充两条曲线之间区域


import matplotlib.pyplot as plt
import numpy as np


n = 1000
x = np.linspace(0, 8*np.pi, n)
sin_x = np.sin(x)
cos_x = np.cos(x/2)/2

plt.figure('filling graph')
plt.xlabel('x')
plt.yla...]]></description><category></category></item><item><title><![CDATA[数据分析2之数据可视化]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103270199</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103270199</guid><author>baidu_38812770</author><pubDate>Wed, 27 Nov 2019 12:03:31 +0800</pubDate><description><![CDATA[1.matplotlib基本使用


import numpy as np
import matplotlib.pyplot as plt

# y=x**2绘制
x_arr = np.arange(9)
y_arr = x_arr**2
plt.plot(x_arr, y_arr)

# 绘制水平线
plt.hlines(30, 2, 7, colors='red')
# 绘制垂直线
# plt...]]></description><category></category></item><item><title><![CDATA[数据分析1之numpy]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/103260567</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/103260567</guid><author>baidu_38812770</author><pubDate>Tue, 26 Nov 2019 17:38:34 +0800</pubDate><description><![CDATA[1.常用库

1. numpy  基础数值算法
2. scipy 科学计算
3. matplotlib 数据可视化
4. pandas 序列高级函数


2.numpy基本使用


import numpy as np
arr = np.array([1, 2, 3, 4])
print(type(arr))

print(ar...]]></description><category></category></item><item><title><![CDATA[链表的插入排序]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/81197659</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/81197659</guid><author>baidu_38812770</author><pubDate>Wed, 25 Jul 2018 10:38:49 +0800</pubDate><description><![CDATA[首先介绍一下直接插入排序，对于无序序列2，4 ，3，1，5。首先初始化两个序列，有序序列{2}和无序序列{4,3,1,5}，把4放进有序序列，从第一个节点开始比较，找到插入位置，把带插入节点插入相应的位置，得到新的序列{2,4}，{3,1,5}，依次进行下去，直至无序序列为空。

下面来介绍链表的插入排序

（1）初始化带头节点L有序链表，ListNode *L=new ListNode(-1)；...]]></description><category></category></item><item><title><![CDATA[二叉树前序中序后序的非递归遍历]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/81196785</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/81196785</guid><author>baidu_38812770</author><pubDate>Wed, 25 Jul 2018 10:16:49 +0800</pubDate><description><![CDATA[1前序遍历

按照根左右的顺序进行遍历，首先访问根节点，若左孩子非空，访问左子树；右孩子非空访问右子树。以此规则遍历整个二叉树

（1）当前点入栈，输出该节点

（2）如当前节点左孩子非空，访问该节点的左孩子回到（1）

（3）若当前节点左孩子为空，访问该节点右孩子回到（1）


    stack&amp;lt;TreeNode*&amp;gt;s;
	vector&amp;lt;int&amp;gt; res;
    Tr...]]></description><category></category></item><item><title><![CDATA[链表中环的入口节点]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/80360992</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/80360992</guid><author>baidu_38812770</author><pubDate>Fri, 18 May 2018 10:53:05 +0800</pubDate><description><![CDATA[题目描述一个链表中包含环，请找出该链表的环的入口结点。设置两个指针，p2每步走两次，p1每步走一次，若它们相遇，快指针比慢指针多走一圈，假定慢指针所走步数为K。头结点pHead到相遇节点步数为k，一圈的步数也为k,有L(pHead,p1)=L(p2,p1),因此L(pHead,p0)=L(p,p0)（对应的优弧的长度）,把p2设置为头结点，同步走，由于p1,p2到环的入口点距离相等，因此p1，p2...]]></description><category></category></item><item><title><![CDATA[二叉搜索树与双向链表]]></title><link>https://blog.youkuaiyun.com/baidu_38812770/article/details/80359813</link><guid>https://blog.youkuaiyun.com/baidu_38812770/article/details/80359813</guid><author>baidu_38812770</author><pubDate>Fri, 18 May 2018 09:48:11 +0800</pubDate><description><![CDATA[题目描述输入一棵二叉搜索树，将该二叉搜索树转换成一个排序的双向链表。要求不能创建任何新的结点，只能调整树中结点指针的指向。首先，二叉搜索树的特点是根&amp;gt;左孩子，根&amp;lt;右孩子。因此，root.left=左子树最右节点，root.right=右子树最左节点。/*
struct TreeNode {
	int val;
	struct TreeNode *left;
	struct TreeNo...]]></description><category></category></item></channel></rss>