\documentclass[10pt,a4paper]{article}

% Packages
\usepackage{fancyhdr}           % For header and footer
\usepackage{multicol}           % Allows multicols in tables
\usepackage{tabularx}           % Intelligent column widths
\usepackage{tabulary}           % Used in header and footer
\usepackage{hhline}             % Border under tables
\usepackage{graphicx}           % For images
\usepackage{xcolor}             % For hex colours
%\usepackage[utf8x]{inputenc}    % For unicode character support
\usepackage[T1]{fontenc}        % Without this we get weird character replacements
\usepackage{colortbl}           % For coloured tables
\usepackage{setspace}           % For line height
\usepackage{lastpage}           % Needed for total page number
\usepackage{seqsplit}           % Splits long words.
%\usepackage{opensans}          % Can't make this work so far. Shame. Would be lovely.
\usepackage[normalem]{ulem}     % For underlining links
% Most of the following are not required for the majority
% of cheat sheets but are needed for some symbol support.
\usepackage{amsmath}            % Symbols
\usepackage{MnSymbol}           % Symbols
\usepackage{wasysym}            % Symbols
%\usepackage[english,german,french,spanish,italian]{babel}              % Languages

% Document Info
\author{cgeeeeh}
\pdfinfo{
  /Title (r.pdf)
  /Creator (Cheatography)
  /Author (cgeeeeh)
  /Subject (R Cheat Sheet)
}

% Lengths and widths
\addtolength{\textwidth}{6cm}
\addtolength{\textheight}{-1cm}
\addtolength{\hoffset}{-3cm}
\addtolength{\voffset}{-2cm}
\setlength{\tabcolsep}{0.2cm} % Space between columns
\setlength{\headsep}{-12pt} % Reduce space between header and content
\setlength{\headheight}{85pt} % If less, LaTeX automatically increases it
\renewcommand{\footrulewidth}{0pt} % Remove footer line
\renewcommand{\headrulewidth}{0pt} % Remove header line
\renewcommand{\seqinsert}{\ifmmode\allowbreak\else\-\fi} % Hyphens in seqsplit
% This two commands together give roughly
% the right line height in the tables
\renewcommand{\arraystretch}{1.3}
\onehalfspacing

% Commands
\newcommand{\SetRowColor}[1]{\noalign{\gdef\RowColorName{#1}}\rowcolor{\RowColorName}} % Shortcut for row colour
\newcommand{\mymulticolumn}[3]{\multicolumn{#1}{>{\columncolor{\RowColorName}}#2}{#3}} % For coloured multi-cols
\newcolumntype{x}[1]{>{\raggedright}p{#1}} % New column types for ragged-right paragraph columns
\newcommand{\tn}{\tabularnewline} % Required as custom column type in use

% Font and Colours
\definecolor{HeadBackground}{HTML}{333333}
\definecolor{FootBackground}{HTML}{666666}
\definecolor{TextColor}{HTML}{333333}
\definecolor{DarkBackground}{HTML}{A3A3A3}
\definecolor{LightBackground}{HTML}{F3F3F3}
\renewcommand{\familydefault}{\sfdefault}
\color{TextColor}

% Header and Footer
\pagestyle{fancy}
\fancyhead{} % Set header to blank
\fancyfoot{} % Set footer to blank
\fancyhead[L]{
\noindent
\begin{multicols}{3}
\begin{tabulary}{5.8cm}{C}
    \SetRowColor{DarkBackground}
    \vspace{-7pt}
    {\parbox{\dimexpr\textwidth-2\fboxsep\relax}{\noindent
        \hspace*{-6pt}\includegraphics[width=5.8cm]{/web/www.cheatography.com/public/images/cheatography_logo.pdf}}
    }
\end{tabulary}
\columnbreak
\begin{tabulary}{11cm}{L}
    \vspace{-2pt}\large{\bf{\textcolor{DarkBackground}{\textrm{R Cheat Sheet}}}} \\
    \normalsize{by \textcolor{DarkBackground}{cgeeeeh} via \textcolor{DarkBackground}{\uline{cheatography.com/191000/cs/39875/}}}
\end{tabulary}
\end{multicols}}

\fancyfoot[L]{ \footnotesize
\noindent
\begin{multicols}{3}
\begin{tabulary}{5.8cm}{LL}
  \SetRowColor{FootBackground}
  \mymulticolumn{2}{p{5.377cm}}{\bf\textcolor{white}{Cheatographer}}  \\
  \vspace{-2pt}cgeeeeh \\
  \uline{cheatography.com/cgeeeeh} \\
  \end{tabulary}
\vfill
\columnbreak
\begin{tabulary}{5.8cm}{L}
  \SetRowColor{FootBackground}
  \mymulticolumn{1}{p{5.377cm}}{\bf\textcolor{white}{Cheat Sheet}}  \\
   \vspace{-2pt}Not Yet Published.\\
   Updated 27th September, 2023.\\
   Page {\thepage} of \pageref{LastPage}.
\end{tabulary}
\vfill
\columnbreak
\begin{tabulary}{5.8cm}{L}
  \SetRowColor{FootBackground}
  \mymulticolumn{1}{p{5.377cm}}{\bf\textcolor{white}{Sponsor}}  \\
  \SetRowColor{white}
  \vspace{-5pt}
  %\includegraphics[width=48px,height=48px]{dave.jpeg}
  Measure your website readability!\\
  www.readability-score.com
\end{tabulary}
\end{multicols}}




\begin{document}
\raggedright
\raggedcolumns

% Set font size to small. Switch to any value
% from this page to resize cheat sheet text:
% www.emerson.emory.edu/services/latex/latex_169.html
\footnotesize % Small font.

\begin{multicols*}{3}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{R语言基础}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{NA: for missing or undefined \seqsplit{data，有这个数字，可是获得不了，比如我的头发（真的很难数嘛？）}} \tn 
% Row Count 3 (+ 3)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{NULL: for empty object (e.g. null / empty \seqsplit{lists)，没有的数字，压根找不到，比如海龟根本没有头发这个概念}} \tn 
% Row Count 6 (+ 3)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{NaN: for results that cannot be reasonably defined} \tn 
% Row Count 7 (+ 1)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{.libPaths() \# get library location； library()   \# see all packages installed； search()    \# see packages currently loaded} \tn 
% Row Count 10 (+ 3)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\# \seqsplit{获取帮助：help.start()；\#} help about function \seqsplit{foo：help(foo)/?foo；exmaple()；args()}} \tn 
% Row Count 12 (+ 2)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\# list all functions containing string foo：apropos("foo") ；\# show an example of function foo：example(foo)} \tn 
% Row Count 15 (+ 3)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{builtins()列出所有的内置函数}} \tn 
% Row Count 16 (+ 1)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\%\% in R = \% in Python；\%/\% inR = // in Python, 不然会返回float; TRUE/T in R = True in Python} \tn 
% Row Count 18 (+ 2)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{\%in\%是in的意思，但是要把list转化成vec。unlist()。再使用}} \tn 
% Row Count 20 (+ 2)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{data \seqsplit{type：integer(5L)/continuous/categorical} (norminal/ordinal)/ text} \tn 
% Row Count 22 (+ 2)
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{matrix里面只能有一个datatype，dt和df没有这个限制}} \tn 
% Row Count 24 (+ 2)
% Row 11
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{options('scipen',100)显示100条向量值} \tn 
% Row Count 25 (+ 1)
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{定义函数：x=function} (x) \{ return (T)\}} \tn 
% Row Count 26 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{\seqsplit{探索性数据分析的心法}}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{1、商业问题是什么}} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{2、我需要知道数据的什么，来帮助理解并回答商业问题，找到商业机会}} \tn 
% Row Count 3 (+ 2)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{3、具体用R怎么实现}} \tn 
% Row Count 4 (+ 1)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{探索性数据分析的目的：}} \tn 
% Row Count 5 (+ 1)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{1、将数据与商业问题结合。数据充足吗？合适吗？比如没有预测性数值，类别变量，冗余}} \tn 
% Row Count 8 (+ 3)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{2、探测数据的问题。数据质量，异常值的监测}} \tn 
% Row Count 10 (+ 2)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{基本语法}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{c("a",1)会返回'a','1'，因为char比numeric更高级，而vec会保证元素的类型全部一样} \tn 
% Row Count 3 (+ 3)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{c(1,2,3)*c(1,2,3)/c(1,2,3)\textasciicircum{}2=c(1,4,9)} \tn 
% Row Count 4 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{class(c(TRUE))返回的是logical}, \seqsplit{class(dt)返回datatable;} \seqsplit{length()返回变量数量，length("TTT")返回1}} \tn 
% Row Count 7 (+ 3)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{str() check classification of \seqsplit{viriables，检查数据框中有哪些数据，包括类型和数值}} \tn 
% Row Count 9 (+ 2)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{summary()可以用于检查数据错误：给出数值型变量的数值summary，min，max，median，mean；给出字符串向量的长度和class；factor向量会给出各个因子的count}} \tn 
% Row Count 13 (+ 4)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{向量，矩阵，df，dt初始化：c()，matrix(data=NA},nrow,ncol)，data.frame("col":vec)，data.table(vec)} \tn 
% Row Count 17 (+ 4)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{向量添加元素：vec \textless{}- append(vec,element)；vec=c(a,a,a)会得到一个flat的向量vec} \tn 
% Row Count 19 (+ 2)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{重命名列：names(df1)} \textless{}-"col"/setnames(df\$col,old=c(xxx),new="xxx")/colnames(dt)=c('1','2','diff')} \tn 
% Row Count 22 (+ 3)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{删除列：col=NULL/df{[}-c(1,2){]}} \tn 
% Row Count 23 (+ 1)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{把两列或者两个别的什么东西粘在一起：col=paste(col1},col2, sep='-')；paste(1,"a",T)} \tn 
% Row Count 26 (+ 3)
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{独特的值：unique(), nrow()} \tn 
% Row Count 27 (+ 1)
% Row 11
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{行列求和：colSums()/rowSums()}} \tn 
% Row Count 28 (+ 1)
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{ifelse语句：ifelse(expression}, 0, 1), a=1; b=if(a==2) 'a' else 'b'} \tn 
% Row Count 30 (+ 2)
\end{tabularx}
\par\addvspace{1.3em}

\vfill
\columnbreak
\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{基本语法 (cont)}}  \tn
% Row 13
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{a=1; if(a==2) b='a' else b='b'} \tn 
% Row Count 1 (+ 1)
% Row 14
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{\#返回对应索引的值：switch(3}, "apple", "orange", "pear", "pineapple") / switch("beta", "alpha"="Big Rock", "beta"="Meteorite",  "gamma"="Red Stones")} \tn 
% Row Count 6 (+ 5)
% Row 15
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{切片索引：by address df{[}c(3:7),seq(1,ncol(df),2){]}} \tn 
% Row Count 8 (+ 2)
% Row 16
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{by value：df{[}(df\textgreater{}=10) \& (df\textless{}=20) {]}} \tn 
% Row Count 9 (+ 1)
% Row 17
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{赋值：df{[}2:4,1{]}=c(3,9,7), df{[}c(3,5), 2:4{]}=rbind(c(1,3,5), c(2,4,6), c(7,9,11))，df{[}1{]}=df{[},1{]}} \tn 
% Row Count 11 (+ 2)
% Row 18
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{批量赋值：df{[}df\textless{}0{]}=-999。所有满足条件的值都被赋一个值} \tn 
% Row Count 13 (+ 2)
% Row 19
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{R语言一直在原数据上更改，而python未必，这就是python会报错要求你去用iloc的原因}} \tn 
% Row Count 16 (+ 3)
% Row 20
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{df{[},2{]}{[}df{[},2{]}\textless{}0{]}=-777；可以先选择列再选择行} \tn 
% Row Count 18 (+ 2)
% Row 21
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{df{[}df{[},1{]}\textless{}0 \& df{[},2{]}\textgreater{}0, {]}=cbind(9,8)} \tn 
% Row Count 19 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{数据预处理}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{数据清洗的目的一是business} object，二是technical requirements} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{新建一列，清洗完之后和原列作比较}} \tn 
% Row Count 3 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{主要目的是防止清洗出错}} \tn 
% Row Count 4 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{\seqsplit{缺失值的处理（删除/填充）}}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{找到缺失的原因，这能决定如何填充缺失值}} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{分类值预测：整列的mode，subgroup的mode，cart，逻辑回归}} \tn 
% Row Count 4 (+ 2)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{连续值预测：随机缺失的值可以用mean填充；系统性缺失的值可以用subgroup的mean，或者cart/逻辑回归；离散化连续值的列，当成分类值处理}} \tn 
% Row Count 8 (+ 4)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{删除一整行只是最后的选择}} \tn 
% Row Count 9 (+ 1)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{logistic/linear模型会自动忽略缺失值的行}} \tn 
% Row Count 11 (+ 2)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{is.na(dt)会显示所有的格子，which是不是空值}} \tn 
% Row Count 13 (+ 2)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{sum(is.na())只会得到一个值}} \tn 
% Row Count 14 (+ 1)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{which(is.na(data))能得到空值在第几行}, \seqsplit{which(h\$counts==max(h\$counts))}} \tn 
% Row Count 16 (+ 2)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{na.omit(ins.dt)} \tn 
% Row Count 17 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{错误值}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{将错误值替换成正确的值}} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{预测错误值，让他们变的更正确}} \tn 
% Row Count 2 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{把它们变成na，交给cart去预测}} \tn 
% Row Count 3 (+ 1)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{删掉一整行} \tn 
% Row Count 4 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{\seqsplit{数据不一致/数据重复}}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{找到数据不一致的原因，尽量从源头解决问题}} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{数据重复的定义，往往取决于主键}} \tn 
% Row Count 3 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{x{2.4885 cm} x{2.4885 cm} }
\SetRowColor{DarkBackground}
\mymulticolumn{2}{x{5.377cm}}{\bf\textcolor{white}{data exploration R}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{\#rm(list=ls())，ls()列出所有的变量}} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{读数据：read.table("csv"}, skip=7, header=T, sep=",", row.names="id",nrows=4)；} \tn 
% Row Count 3 (+ 2)
% Row 2
\SetRowColor{LightBackground}
read.csv("csv", stringsAsFactors = \seqsplit{TRUE)。stringsAsFactors把字符串转化为因子变量} & fread("csv",, na.strings = c("NA", "missing", "N/A", -99, "", "m", "M", "na", ".")) \tn 
% Row Count 8 (+ 5)
% Row 3
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{写数据：write.csv(df,path, row.names=F); write.table(df,path)} \tn 
% Row Count 10 (+ 2)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{rbind(df,df2), cbind(df.df2), names(df){[}3:4{]}=c('a','b')} \tn 
% Row Count 12 (+ 2)
% Row 5
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{查看导入的数据集：head(df)},view(df), 图形交互} \tn 
% Row Count 14 (+ 2)
% Row 6
\SetRowColor{LightBackground}
\seqsplit{数据维度：dim(df)} & \seqsplit{数据列的细节，如有多少缺失值，各个类别有几个数：summary(df)} \tn 
% Row Count 19 (+ 5)
% Row 7
\SetRowColor{white}
\seqsplit{查看子集：subset(df}, is.na(df\$col, \seqsplit{select))。select选中了数据集中的某列} & train \textless{}- sample.split(Y = mtcars\$mpg, SplitRatio = 0.7) trainset \textless{}- subset(mtcars, train == T) testset \textless{}- subset(mtcars, train == F) \tn 
% Row Count 26 (+ 7)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{比较两个向量或者df是否相同：identical(df1},df2)} \tn 
% Row Count 28 (+ 2)
% Row 9
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{为数据集增加噪音：jitter(df\$col)，jitter只能给连续变量加噪音，logical和类别变量不行}} \tn 
% Row Count 31 (+ 3)
\end{tabularx}
\par\addvspace{1.3em}

\vfill
\columnbreak
\begin{tabularx}{5.377cm}{x{2.4885 cm} x{2.4885 cm} }
\SetRowColor{DarkBackground}
\mymulticolumn{2}{x{5.377cm}}{\bf\textcolor{white}{data exploration R (cont)}}  \tn
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{\#生成sequence，重复数字，反转列表：seq(10},20,3)，rep(10,5)，rev(v),sort(v)升序,unique(v),} \tn 
% Row Count 3 (+ 3)
% Row 11
\SetRowColor{white}
vector(c, n) – returns vector with all values c of size n & names(vec)=c('a','b','c')可以给vec的元素取名字 \tn 
% Row Count 6 (+ 3)
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{matrix(c, nrow=5, ncol=3) – returns a 53 matrix with all values c} \tn 
% Row Count 8 (+ 2)
% Row 13
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{data.frame(v1, v2, v3…) – returns a data frame made up of column  vectors v1, v2, v3,} \tn 
% Row Count 10 (+ 2)
% Row 14
\SetRowColor{LightBackground}
\seqsplit{所以要转换类别：as}.numeric(df\$col) & \seqsplit{as.integer()=int()in} python \tn 
% Row Count 13 (+ 3)
% Row 15
\SetRowColor{white}
\seqsplit{对每一列都执行某样操作：sapply(df}, func, \seqsplit{na.rm=TRUE)。Possible} functions used in sapply include mean, sd, var, min, max, median, range , and quantile & sapply(v, function(x) if (is.na(x)) 999 else x) \tn 
% Row Count 22 (+ 9)
% Row 16
\SetRowColor{LightBackground}
lapply(my\_list, function(x) x == element)) & \seqsplit{lapply和sapply的区别是返回结构不同，lapply返回list，sapply返回向量和矩阵} \tn 
% Row Count 27 (+ 5)
% Row 17
\SetRowColor{white}
\seqsplit{\#见堆叠柱状图：table(deparse}.level=2)/prop.table(col1,col2,margin)能把两列数据组合成透视表 & \seqsplit{deparse.level=2会展示所有行和列的名字} \tn 
% Row Count 33 (+ 6)
\end{tabularx}
\par\addvspace{1.3em}

\vfill
\columnbreak
\begin{tabularx}{5.377cm}{x{2.4885 cm} x{2.4885 cm} }
\SetRowColor{DarkBackground}
\mymulticolumn{2}{x{5.377cm}}{\bf\textcolor{white}{data exploration R (cont)}}  \tn
% Row 18
\SetRowColor{LightBackground}
\seqsplit{所有的factor：levels(factorcol}, ordered=T, levels=c(xxx), labels=c(xxx)) & \seqsplit{labels把向量里的数值映射到一个新的空间，它和levels的区别是levels只涉及input，labels涉及到output} \tn 
% Row Count 7 (+ 7)
% Row 19
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{relevel(x, \seqsplit{ref)将ref因子放在x的第一个}} \tn 
% Row Count 8 (+ 1)
% Row 20
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{\seqsplit{切分左开右闭区间：limits=c(1，2，3，4，5)}} \tn 
% Row Count 10 (+ 2)
% Row 21
\SetRowColor{white}
\mymulticolumn{2}{x{5.377cm}}{生成新的一列，值是区间：dt{[}, new\_col := cut(col, breaks=limits, include.lowest=T){]}} \tn 
% Row Count 12 (+ 2)
% Row 22
\SetRowColor{LightBackground}
\seqsplit{cor计算相关系数，the} degree of the consistensy of the trend of the relationship & \seqsplit{解读时了一说，一个上升时，另一个倾向于xxx，但因果关系我们并不清楚} \tn 
% Row Count 17 (+ 5)
% Row 23
\SetRowColor{white}
\seqsplit{cor(dataframe)可以计算相关性矩阵，"corrplot"；corrplot(cor(mtcars)}, type = "upper") & \seqsplit{cov计算协方差，与cor（相关系数不同）} \tn 
% Row Count 22 (+ 5)
% Row 24
\SetRowColor{LightBackground}
\mymulticolumn{2}{x{5.377cm}}{ceiling(2.5)=3} \tn 
% Row Count 23 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}--}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{随机数}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{var(), sd()（标准差）, skewness(),kurtosis()} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{quantile(data,c(0.025,0.975))} \tn 
% Row Count 2 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{sample(1:80,80,replace=F)；第一个是参数，第二个是size，第三个是能否重复} \tn 
% Row Count 4 (+ 2)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{dbinom(x,n,p)，pbinom。x次成功from \seqsplit{n次，每次概率为p。x可以是1：10，成功1：10次}} \tn 
% Row Count 6 (+ 2)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{dpois(x,lambda), ppois(x,lambda)} \tn 
% Row Count 7 (+ 1)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{dnorm(x,mean,sd)，pnorm(x,mean,sd)} \tn 
% Row Count 8 (+ 1)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{runif(size, low, high), rnorm(size, mean, sd而不是方差), rbinom(size, times, possibility)} \tn 
% Row Count 10 (+ 2)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{set.seed(seed) fixes the random result of ramdom function} \tn 
% Row Count 12 (+ 2)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{curve(dnorm(x,mean=0,sd=1),col="red",from=-3,to=4,xlim=c(-3,4),ylim=c(0,1)); curve(dnorm(x,mean=1,sd=1),col="blue",add=T)} \tn 
% Row Count 15 (+ 3)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{t.test(x, y = NULL, alternative = c("two.sided", "less", "greater"),        mu = 0, paired = FALSE, var.equal = FALSE,        conf.level = 0.95)} \tn 
% Row Count 18 (+ 3)
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{如果x也有值，y也有值，那就是计算x-y的置信区间}} \tn 
% Row Count 20 (+ 2)
% Row 11
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{prop.test(x, n, p = NULL,           alternative = c("two.sided", "less", "greater"),           conf.level = 0.95, correct = TRUE)} \tn 
% Row Count 23 (+ 3)
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{这里x可以是向量，两次trial的成功次数；n是总次数，同理。算出来就是做差值}} \tn 
% Row Count 26 (+ 3)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{ggplot2}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{data层：ggplot(data,aes(x=colname,y=colname, fill=factorcolname))} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{图像层：geom\_point()} \tn 
% Row Count 3 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{布局层：facet\_grid(.\textasciitilde{}fl)。.\textasciitilde{}fl是指分组的colname} \tn 
% Row Count 5 (+ 2)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{将factor变量映射到颜色：scae\_fill\_manual(values=c("0"="dark} blue", "1"="orange")} \tn 
% Row Count 7 (+ 2)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{细节层：labs(title, xlab,ylab)} \tn 
% Row Count 8 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{visualization with R}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{par(mfrow=c(2,2))，两行两列的子图} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{plot(xaxt="n",yaxt="n");抹掉所有坐标轴} \tn 
% Row Count 2 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{axis(1,at=seq(0.5,length(),1),labels=names(),tick=F,col="red")1是x轴，at决定位置，labels决定具体显示，包括内容和间距，tick控制有没有坐标线，col决定标线颜色} \tn 
% Row Count 6 (+ 4)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{label太多，可调整字体方向：par(las=0},1,2,3)；0=parallel, 1=all horizontal, 2=all perpendicular to axis, 3=all vertical} \tn 
% Row Count 9 (+ 3)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{图片太大，可调整图与周边margin：par(mar=c(5},4,4,2)+0.1)，margin的顺序是下，左，上，右；单位是line} \tn 
% Row Count 12 (+ 3)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{允许画图画到外面去：par(xpd=T)。}} \tn 
% Row Count 13 (+ 1)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{添加legend：legend("topright"}, inset=c(0,0), fill=c("red","grey"), legend=rownames(counts), border="grey", cex=0.6)} \tn 
% Row Count 16 (+ 3)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{第一个是大体位置，inset是具体位置，fill是对对应的颜色，legend是图例名字，border是边框颜色，}} \tn 
% Row Count 19 (+ 3)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{散点图和箱线图的区别是：散点图可以告诉你样本量的大小（比如50岁左右的人有保险的比没有的多），而箱线图不行}} \tn 
% Row Count 23 (+ 4)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{plot(density(df\$col), xlab,ylab,main)} \tn 
% Row Count 24 (+ 1)
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{hist(df\$col, ylim=c(0,220), breaks=c(-10,0,10,20), labels=T, col="light blue") \seqsplit{labels会给每个柱子加上数字；默认每个区间是左开右闭的}} \tn 
% Row Count 28 (+ 4)
% Row 11
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{boxplot(df\$col \textasciitilde{} df\$catecol)；use  \seqsplit{\$stats查看两个箱线图计算出的几个critical数据，从上到下依次递增的数据}} \tn 
% Row Count 31 (+ 3)
\end{tabularx}
\par\addvspace{1.3em}

\vfill
\columnbreak
\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{visualization with R (cont)}}  \tn
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{箱线图的数据依次是：Q1-1}.5IQR，Q1，median，Q3，Q3+1.5QIR。inter-quatile \seqsplit{range简称IQR。在box-and-whisker} \seqsplit{method里，其余的都叫做异常值}} \tn 
% Row Count 4 (+ 4)
% Row 13
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{柱状图，展示类别变量的分布：barplot(table(df\$col)},col=c("light blue","mistyrose","lightcyan","lavander"), horiz=T, cex.names=0.5)。} \tn 
% Row Count 7 (+ 3)
% Row 14
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{堆叠柱状图：data=table(df\$ycol},df\$xcol)。row \seqsplit{index是ycol的各个数值，col} \seqsplit{index是xcol的各个值。barplot(data}, col=c("red","grey"))} \tn 
% Row Count 10 (+ 3)
% Row 15
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{百分比柱状图(主要是数据预处理的不同)：prop}.table(df\$ycol,df\$xcol, \seqsplit{margin=1/2)。margin=1是横着计算百分比，2是竖着计算百分比}} \tn 
% Row Count 14 (+ 4)
% Row 16
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{散点图：plot(df\$xcol1},df\$ycol2)； \seqsplit{在散点基础上加一根smooth} \seqsplit{curve。scatter.smooth(df\$col1},df\$col2, \seqsplit{col="grey")；col决定散点的颜色}} \tn 
% Row Count 18 (+ 4)
% Row 17
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{散点曲线矩阵图，查看各个变量间的关系，分辨哪个最先分析：pairs(\textasciitilde{} col1+age+sex+...., panel=panel.smooth, span=0.75, \seqsplit{data=df)。panel.smooth是加上平滑曲线，span越大，线性程度越高}} \tn 
% Row Count 23 (+ 5)
% Row 18
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{png(),jpeg()+dev.off()能存储图片} \tn 
% Row Count 24 (+ 1)
% Row 19
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{画完图，sys.sleep(0.05)}} \tn 
% Row Count 25 (+ 1)
% Row 20
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{绘制空白图} \tn 
% Row Count 26 (+ 1)
% Row 21
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{plot(x = c(22, 28), y = c(1, 1000), type = "n", xlab = "",      ylab = "") \# set up a blank plot with specified ranges} \tn 
% Row Count 29 (+ 3)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{data.table}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{data.table中，j参数里，：=的结果within在dt中，=在out新建一个dt}} \tn 
% Row Count 2 (+ 2)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{创建data.table：dt{[}, .(.N, colname=sum(col==2), \seqsplit{prop.uninsured=sum(col==2)/.N)}, keyby=colname{]}} \tn 
% Row Count 4 (+ 2)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{keyby和by的区别是：keyby会sorting分组的结果。可以有多个分组标准}.(col1,col2)} \tn 
% Row Count 6 (+ 2)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{.N是指列名是number，值是count}} \tn 
% Row Count 7 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{线性回归}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{多重共线性不影响预测，只影响模型解读。同时解读模型时要假设其他变量不变。不能反写成x的等式，因为这不是代数，是统计模型}} \tn 
% Row Count 4 (+ 4)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{assumption1：linear assosiation between y and x} \tn 
% Row Count 5 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{assumption2：error has a normal distribution with mean 0} \tn 
% Row Count 7 (+ 2)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{assumption3：errors与x互相独立，并且有常数的standard} deviation} \tn 
% Row Count 9 (+ 2)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{lm(y\textasciitilde{}x1+x2或者., data=data)；m4 \textless{}- \seqsplit{step(m.full);赤潮信息准则}} \tn 
% Row Count 11 (+ 2)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{coef(model); confint(model)} \tn 
% Row Count 12 (+ 1)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{abline(m1, col = "red")；identify(x = mtcars\$wt, y = mtcars\$mpg)} \tn 
% Row Count 14 (+ 2)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{win.graph()；identify(x = x\_data, y = y\_data)} \tn 
% Row Count 15 (+ 1)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{R \seqsplit{suqare代表了模型的解释力。about} xxx\% of the data canbe explained bythe \seqsplit{model。只要增加变量数，R方会一直上升，因此不能简单用R方来比较两个线性模型}} \tn 
% Row Count 19 (+ 4)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{adjusted R \seqsplit{square：惩罚每一个被添加的变量。在多变量的前提下，用这个比r方要好}} \tn 
% Row Count 22 (+ 3)
% Row 10
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{plot（lm函数的结果）来检验假设1，2，3}} \tn 
% Row Count 24 (+ 2)
% Row 11
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{左上角残差图：test1，2。理想情况是y=0的一条红线。residuals(m5)}} \tn 
% Row Count 26 (+ 2)
% Row 12
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{右上角qq图：test2.理想情况是沿着虚线}} \tn 
% Row Count 27 (+ 1)
% Row 13
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{左下角经过标准化后的残差图：test3。理想情况是上下均匀地分布在一个矩形里，而不是随着x的增大而改变}} \tn 
% Row Count 30 (+ 3)
\end{tabularx}
\par\addvspace{1.3em}

\vfill
\columnbreak
\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{线性回归 (cont)}}  \tn
% Row 14
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{右下角：展示influential} \seqsplit{outliers。有木有influence主要指去掉这个点对拟合曲线的影响有多大}} \tn 
% Row Count 3 (+ 3)
% Row 15
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{单变量离群点很好辨认，超过两个变量散点图就不能用了，所以得用cook统计量，点落到在虚线外就是influential}} \tn 
% Row Count 6 (+ 3)
% Row 16
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{处理离散类别变量，要注意用r转换成factor。然后哪怕是有序变量，数字本身也没有意义，不能当成连续变量来处理}} \tn 
% Row Count 10 (+ 4)
% Row 17
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{k个类，就有k-1个变量，其中一个会是baseline，baseline自动是字母顺序表里最早的那个，然后也可以用relevel（）自定义}} \tn 
% Row Count 14 (+ 4)
% Row 18
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{如何决定选择哪些变量：}} \tn 
% Row Count 15 (+ 1)
% Row 19
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{1、专家，领域知识}} \tn 
% Row Count 16 (+ 1)
% Row 20
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{2、统计知识：pvalue小于5\%},前向，后向选择，双向选择，降维方法，CRT等} \tn 
% Row Count 18 (+ 2)
% Row 21
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{多重共线性：一个x能被其他x线性表出，意味着这个x的信息被其他的包含进去了。因此dummy} variable要减去1} \tn 
% Row Count 21 (+ 3)
% Row 22
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{vif=1/（1-Ri\textasciicircum{}2）。Ri\textasciicircum{}2就是以这个x为因变量，其他x为自变量回归得出的R2。vif(lm的return)} \tn 
% Row Count 24 (+ 3)
% Row 23
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{一版vif\textgreater{}5或者10，有dummy variable的模型一版gvif\textgreater{}2。使用vif模型from package car} \tn 
% Row Count 26 (+ 2)
% Row 24
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{预测未来值：predict}.m5.test \textless{}- predict(m5, newdata = testset, type='response')response 返回y=1的概率值，} \tn 
% Row Count 29 (+ 3)
% Row 25
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{提取p值：summary(model)\$coefficients{[},4{]}} \tn 
% Row Count 30 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{\seqsplit{逻辑回归（预测分类变量）}}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{glm(y\textasciitilde{}x, family=binomial, data=data)} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{1、建模预测分类变量}} \tn 
% Row Count 2 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{2、如何辨别高风险因子}} \tn 
% Row Count 3 (+ 1)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{3、双变量分类模型}} \tn 
% Row Count 4 (+ 1)
% Row 4
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{4、odds（胜利/失败概率，chances），在代数中等于e\textasciicircum{}z,是个function，odds ratio代表每个bk的作用，e\textasciicircum{}bk，是个常数} \tn 
% Row Count 7 (+ 3)
% Row 5
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{连续变量xk增加一个单位，胜率会怎么增加，会乘以e\textasciicircum{}bk} \tn 
% Row Count 9 (+ 2)
% Row 6
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{类别变量xk从baseline跳转到一个类，胜率也会乘以e\textasciicircum{}bk} \tn 
% Row Count 11 (+ 2)
% Row 7
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{判断一个xk变化会怎么影响odds} \seqsplit{ratio，可以用置信区间，2.5\%-97.5\%的区间超过1且不包含1则大于1}} \tn 
% Row Count 14 (+ 3)
% Row 8
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{\seqsplit{5、multinomial（超过3个类别）}} \tn 
% Row Count 15 (+ 1)
% Row 9
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{multinom() function from nnet Rpackage} \tn 
% Row Count 16 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}

\begin{tabularx}{5.377cm}{X}
\SetRowColor{DarkBackground}
\mymulticolumn{1}{x{5.377cm}}{\bf\textcolor{white}{cluster}}  \tn
% Row 0
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{library(cluster)} \tn 
% Row Count 1 (+ 1)
% Row 1
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{km2=kmeans(pts,centers=center)\#初始的中心} \tn 
% Row Count 2 (+ 1)
% Row 2
\SetRowColor{LightBackground}
\mymulticolumn{1}{x{5.377cm}}{clus1=pts{[}km2\$cluster==1{]}} \tn 
% Row Count 3 (+ 1)
% Row 3
\SetRowColor{white}
\mymulticolumn{1}{x{5.377cm}}{agnes(rivers),plot(that)} \tn 
% Row Count 4 (+ 1)
\hhline{>{\arrayrulecolor{DarkBackground}}-}
\end{tabularx}
\par\addvspace{1.3em}


% That's all folks
\end{multicols*}

\end{document}