☰

! 额外 }，或忘记 $

tag-icon • tag-icon algorithms
! 额外 }，或忘记 $

以下算法有什么问题？

\documentclass[journal]{IEEEtran}
 
\usepackage{graphicx}
\usepackage{subfigure}
\usepackage{overpic}
\usepackage{amsmath}
\usepackage{flushend}
\usepackage{amssymb}
\usepackage{multirow}
\usepackage{makecell}
\usepackage{color, soul}
\usepackage[ruled]{algorithm2e}


\begin{algorithm}
\caption{Sample Mixup Algorithm}\label{al_sm}
\LinesNumbered
\KwIn{Training dataset $(X,Y)$, number of training epochs $m$, two parameters $p$ and $q$ satisfying (0 $\leq$ $p$ $<$ $q$ $\leq$ 1), Beta distribution parameter $\alpha$ for sample mixup.}
\For{$i = 1$ $to$ $m$}{
Draw a mini-batch ($x_b$, $y_b$).\\
\uIf(\tcc*[f]{First stage}){$i$ $\leq$ $pm$}{
($\Tilde{x_b}$, $\Tilde{y_b}$) = mixup($x_b$, $y_b$, $\alpha$)}
\uElseIf(\tcc*[f]{Second stage}){$i$ $\leq$ $qm$}{
$\epsilon = \frac{m-i}{m(1-q)}$\\
Randomly generate threshold $\theta\in \mathbb{$[0,1]$}$.\\
\uIf{$\theta < \epsilon$}{
($\Tilde{x_b}$, $\Tilde{y_b}$) = mixup($x_b$, $y_b$, $\alpha$)}
\Else{($\Tilde{x_b}$, $\Tilde{y_b}$) = ($x_b$, $y_b$)}}
\Else(\tcc*[f]{Third stage}){($\Tilde{x_b}$, $\Tilde{y_b}$) = ($x_b$, $y_b$)}
Train model with mini-batch ($\Tilde{x_b}$, $\Tilde{y_b}$).}
\end{algorithm}

答案1

改编

  • 添加document环境
  • 错误消息的原因（! Extra }, or forgotten $）：更改\mathbb{$[0,1]$}为[0,1]（不要再在数学模式中使用数学模式）
  • 使用缩进以避免丢失代码（和花括号）
  • 不要把每一个符号放在单独的数学环境中，例如，0 $\leq$ $p$ $<$ $q$ $\leq$ 1应该是$0 \leq p < q \leq 1$
  • 为 MWE 提供折扣套餐

代码

\documentclass[journal]{IEEEtran}
 
\usepackage{amsmath}
\usepackage[ruled]{algorithm2e}

\begin{document}

\begin{algorithm}
    \caption{Sample Mixup Algorithm}
    \label{al_sm}
    \LinesNumbered
    \KwIn{Training dataset $(X,Y)$, number of training epochs $m$, two parameters $p$ and $q$ satisfying ($0 \leq p < q \leq 1$), Beta distribution parameter $\alpha$ for sample mixup.}
    \For{$i = 1$ to $m$}{
        Draw a mini-batch ($x_b$, $y_b$).\\
        \uIf(\tcc*[f]{First stage}){$i \leq pm$}{
            ($\Tilde{x_b}$, $\Tilde{y_b}$) = mixup($x_b$, $y_b$, $\alpha$)
        } \uElseIf(\tcc*[f]{Second stage}){$i$ $\leq$ $qm$}{
            $\epsilon = \frac{m-i}{m(1-q)}$\\
            Randomly generate threshold $\theta\in [0,1]$.\\
            \uIf{$\theta < \epsilon$}{
                ($\Tilde{x_b}$, $\Tilde{y_b}$) = mixup($x_b$, $y_b$, $\alpha$)
            } \Else{
                ($\Tilde{x_b}$, $\Tilde{y_b}$) = ($x_b$, $y_b$)
            }
        } \Else(\tcc*[f]{Third stage}){
            ($\Tilde{x_b}$, $\Tilde{y_b}$) = ($x_b$, $y_b$)
        }
        Train model with mini-batch ($\Tilde{x_b}$, $\Tilde{y_b}$).
    }
\end{algorithm}

\end{document}

相关内容