Latex 中进行算法的排版

在论文写作和排版过程中，常常会用到算法描述，在LaTex中，算法描述块的排版会用到两个宏包 \usepackage{algorithm}  和  \usepackage{algorithmic}。算法的排版，主要在于控制缩进、粗体、横线等格式，这些都会在这篇博客中进行介绍。

在开始算法排版之前，首先在文档开头加入下面两句，以导入宏包：

\usepackage{algorithm}  
\usepackage{algorithmic}  

• 例1
• 例2
• 例3
• 例4
• 例5

例1

那么，我们首先看一个例子：

\begin{algorithm}  
\caption{A}  
\label{alg:A}  
\begin{algorithmic}  
\STATE {set $r(t)=x(t)$}   
\REPEAT   
\STATE set $h(t)=r(t)$   
\REPEAT  
\STATE set $h(t)=r(t)$   
\UNTIL{B}   
\UNTIL{B}  
\end{algorithmic}  
\end{algorithm}  

在编译之后，显示为：

使用algorithmic包时，关键字全部大写，如果使用的是algorithmicx包，那么关键字首字母大写，后面小写。

例2

第二个例子更加详细的展示了缩进的控制，可以自己编译一下：

\begin{algorithm}  
\caption{Calculate $y = x^n$}   
\label{alg1}  
\begin{algorithmic}  
\REQUIRE $n \geq 0 \vee x \neq 0$   
\ENSURE $y = x^n$   
\STATE $y \Leftarrow 1$   
\IF{$n < 0$}   
\STATE $X \Leftarrow 1 / x$   
\STATE $N \Leftarrow -n$   
\ELSE   
\STATE $X \Leftarrow x$   
\STATE $N \Leftarrow n$  
\ENDIF   
\WHILE{$N \neq 0$}   
\IF{$N$ is even}   
\STATE $X \Leftarrow X \times X$   
\STATE $N \Leftarrow N / 2$   
\ELSE[$N$ is odd]   
\STATE $y \Leftarrow y \times X$   
\STATE $N \Leftarrow N - 1$   
\ENDIF   
\ENDWHILE  
\end{algorithmic}  
\end{algorithm}

例3

如果需要显示Input和Output：

\begin{algorithm}  
\caption{Fourier-Mellin Based KCF}  
\label{alg:A}  
\hspace*{0.02in}{\bf Input:}
Image $I$\\preprocessed kernelized template $T_\kappa$\\
\hspace*{0.02in}{\bf Output:} 
 scale $\sigma$, angle $\theta$ relation between $I$ and $T$ 

\begin{algorithmic}[1] 
\STATE {fourier transform: $F=\mathcal{F}(I)$}
\STATE {high pass filter: $F_h=\mathcal{H}(F)$\\$\mathcal{H}(x,y)=(1.0-cos(\pi x)cos(\pi y))(2.0-cos(\pi x)cos(\pi y))$}
\STATE {log-polar transform: $F_{lp}=\mathcal{L}(F_h)$}
\STATE {apply kernel function: $F_\kappa=\mathcal{K}(F_{lp})$}
\STATE {phase correlation: $(\Delta x, \Delta y)=\mathcal{C}(F_\kappa, T_\kappa)$}
\STATE {resolove scale and rotation:\\
$\theta=\alpha \Delta x$, $\sigma=log(\Delta y)$\\
where $\alpha$ is translation factor of pixel translation on fourier domain and polar angle on origin image
}
\end{algorithmic}  
\end{algorithm}

这样，就在开头显示了输入和输出。{algorithmic}[1]表示显示行号，当然，还可以显示竖线，不过要使用额外的宏包，请参考文后链接。

例4

还可以使用\renewcommand 改变现有命令，在导言区加入下列语句

\renewcommand{\algorithmicrequire}{ \textbf{Input:}} %Use Input in the format of Algorithm  
\renewcommand{\algorithmicensure}{ \textbf{Output:}} %UseOutput in the format of Algorithm 

使得原来软件包中定义的命令\REQUIRE和\ENSURE显示为Input:和Output:

\begin{algorithm}[htb]   
\caption{ Framework of ensemble learning for our system.}   
\label{alg:Framwork}   
\begin{algorithmic}[1] %这个1 表示每一行都显示数字  
\REQUIRE ~~\\ %算法的输入参数：Input  
The set of positive samples for current batch, $P_n$;\\  
The set of unlabelled samples for current batch, $U_n$;\\  
Ensemble of classifiers on former batches, $E_{n-1}$;  
\ENSURE ~~\\ %算法的输出：Output  
Ensemble of classifiers on the current batch, $E_n$;  
\STATE Extracting the set of reliable negative and/or positive samples $T_n$ from $U_n$ with help of $P_n$;   
\label{ code:fram:extract }%对此行的标记，方便在文中引用算法的某个步骤  
\STATE Training ensemble of classifiers $E$ on $T_n \cup P_n$, with help of data in former batches;   
\label{code:fram:trainbase}  
\STATE $E_n=E_{n-1}\cup E$;   
\label{code:fram:add}  
\STATE Classifying samples in $U_n-T_n$ by $E_n$;   
\label{code:fram:classify}  
\STATE Deleting some weak classifiers in $E_n$ so as to keep the capacity of $E_n$;   
\label{code:fram:select}  
\RETURN $E_n$; %算法的返回值  
\end{algorithmic}  
\end{algorithm}  

排版结果如下：

例5

最后一个例子：

\begin{algorithm}[h]  
\caption{An example for format For \& While Loop in Algorithm}  
\begin{algorithmic}[1]  
\FOR{each $i \in [1,9]$}  
\STATE initialize a tree $T_{i}$ with only a leaf (the root);\  
\STATE $T=T \cup T_{i};$\  
\ENDFOR  
\FORALL {$c$ such that $c \in RecentMBatch(E_{n-1})$}   
\label{code:TrainBase:getc}  
\STATE $T=T \cup PosSample(c)$;   
\label{code:TrainBase:pos}  
\ENDFOR  
\FOR{$i=1$; $i<n$; $i++$ }  
\STATE $//$ Your source here;  
\ENDFOR  
\FOR{$i=1$ to $n$}  
\STATE $//$ Your source here;  
\ENDFOR  
\STATE $//$ Reusing recent base classifiers.   
\label{code:recentStart}  
\WHILE {$(|E_n| \leq L_1 )and( D \neq \phi)$}  
\STATE Selecting the most recent classifier $c_i$ from $D$;  
\STATE $D=D-c_i$;  
\STATE $E_n=E_n+c_i$;  
\ENDWHILE   
\label{code:recentEnd}  
\end{algorithmic}  
\end{algorithm}  

排版结果为：

 内容参考博主