问题：algorithm2e 包如何使用

[ChenDRAG]

我使用最新版本的模板，在thusetup.txt里面添加了
\usepackage[ruled, vlined]{algorithm2e}

而后在data文件夹下面某个章节插入伪代码（这个伪代码是我原封不动从自己的英文论文里复制来的所以大概率没问题）

\begin{algorithm}[H]
\label{trainer}
% \SetAlgoNoLine
\SetAlgoLined
\LinesNumbered
\SetNlSty{textbf}{}{:}
% \KwResult{Write here the result }
\SetKwFor{RepTimes}{repeat}{times}{end}
\KwIn{n-epochs $N$, steps-per-epoch $L$, steps-per-collect $l$, updates-per-step $u$}
% \KwIn{n-epochs $N$, steps-per-epoch $L$, steps(episodes)-per-collect $l$, updates-per-step $u$}
\KwData{policy, collector, test\_collector, logger.}
    \RepTimes{N}{
      T = 0\;
      \While{$T < L$}{
      Collect l steps by running \emph{result = collector.collect(n\_step(episode) = $l$)}, training data goes to \emph{collector.buffer} and statistics to \emph{result}\;
      
      \emph{T += result['n\_steps']} because we might not exactly collect l steps\;
      
      \emph{logger.log\_train\_data(result)} at regular interval\;
      
        %   \emph{result = collector.collect(n\_step(episode) = $l$)} will collect l timesteps into buffer\;
      \RepTimes{round($u*$result['n\_steps'])}{
        Update policy: \emph{losses = policy.update(collector.buffer)}\;
        \emph{logger.log\_update\_data(losses)} at regular interval\;}
        \label{clearnow}
        }
    Evaluate: \emph{logger.log\_test\_data(test\_collector.collect(n\_episode= \#test\_seeds))}\;
    }
 \caption{Off-policy Trainer}
\end{algorithm}

然而渲染出来就报错了，多为
undifined control sequence,

长这个样

感觉应该是我引用包的方式不对，求教~~~

[ChenDRAG]

额，似乎暂时解决了，需要注释掉thusetup.tex里面

\usepackage{algorithm}
\usepackage{algorithmic}

这两行，似乎是包有冲突之类的吧，我也不太懂这几个包分别有什么用。

[zepinglee]

algorithm 跟 algorithm2e 冲突

[ChenDRAG]

algorithm 跟 algorithm2e 冲突

请问这三个包有啥区别嘛，一直搞不懂是干啥的233

[zepinglee]

algorithm 跟 algorithm2e 冲突

请问这三个包有啥区别嘛，一直搞不懂是干啥的233

algorithm + algorithmic 和 algorithm2e 功能相近，前者较早，后者较新。具体参考各自的文档。