\documentclass{article}
\hyphenation{op-tical net-works semi-conduc-tor}
\usepackage{csquotes}
\usepackage{graphicx}
\usepackage{mathtools}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{mathtools}
\usepackage{lipsum}
\DeclareMathOperator{\RelU}{RelU}
%\usepackage{url}
%\usepackage{algorithm}
%\usepackage{array}
\usepackage{subfloat}
%\usepackage{subfig}
\usepackage{xcolor}
\usepackage{longtable}
%\usepackage{subcaption}
\usepackage{epstopdf}
\usepackage[utf8]{inputenc}
\usepackage[justification=centering]{caption}
%\usepackage{booktabs}
\newcommand{\dd}[1]{\mathrm{d}#1}
%\usepackage{amsmath}
\usepackage{graphicx}
\usepackage[utf8]{inputenc}
\usepackage[english]{babel}
\usepackage{url}
%\usepackage{multicol}
\usepackage{multirow}
\usepackage{array}
\usepackage{cleveref}
\usepackage{calc}
\usepackage[english]{babel}
%\usepackage[document]{ragged2e}
\usepackage{algorithm}
\usepackage{algpseudocode}
\usepackage{subcaption}
\usepackage{booktabs}
\usepackage{wrapfig}
\begin{document}
\title{ A deep learning approach}
\author{Ghul$^{1,2}$, Javaid$^{1*}$, \\
$^{1}$COMSATS University Islamabad 44000, Pakistan\\
$^{2}$University of Engineering and Technology, Mardan, 23200, Pakistan\\
%$^{2}$CAMS, Dept of Biomedical Technology, KSU, Riyadh 11633, Saudi Arabia\\
%$^{3}$FE, Dalhousie University, Halifax, NS B3J 4R2, Canada\\
$^{*}$Corresponding author: www.njavaid.com, [email protected]
}
\maketitle
\begin{abstract}
\end{abstract}
\begin{IEEEkeywords}
Short term load forecasting; deep learning; smart grid; factored conditional restricted boltzmann machine; conditional restricted boltzmann machine; rectified linear unit.
\end{IEEEkeywords}
\section*{Nomenclature}
\addcontentsline{toc}{section}{Nomenclature}
\begin{IEEEdescription}[\IEEEusemathlabelsep\IEEEsetlabelwidth{$V_1,V_2,V_3$}]
\item[$E$] {Error function}
\item[$v$] {Visible layer}
\item[$u$] {History layer}
\item[$h$] Hidden layer
\item[$a$] Visible layer bias
\item[$b$] Hidden layer bias
\item[$w^{vh}$] Bi-directional weight matrix between hidden and visible layers
\item[$w^{uh}$] Unidirectional weight matrix between hidden and visible layers
\item[$w^{uv}$] Unidirectional weight matrix between visible and history layers
\item[$Sigmoid$] Sigmoidal activation function
%\item[$RelU$] Rectified linear unit activation function
\item[$w^{vh}_{t+1}$] Updated bi-directional weight matrix between hidden and visible layers
\item[$w^{uh}_{t+1}$] Updated unidirectional weight matrix between hidden and visible layers
\item[$w^{uv_{t+1}}$] Updated unidirectional weight matrix between visible and history layers
\item[$a_{t+1}$] Updated bias of visible layer
\item[$b_{t+1}$] Updated bias of hidden layer
\item[$\hat a$] Visible layer dynamic bias
\item[$\hat b$] Hidden layer dynamic bias
\item[$w^v$] Visible layer weight
\item[$w^y$] Style layer weight
\item[$w^h$] Hidden layer weight
\item[$A^u$] History layer connection for dynamic bias $\hat a$
\item[$A^v$] Visible layer connection for dynamic bias $\hat a$
\item[$A^y$] Style layer connection for dynamic bias $\hat a$
\item[$B^u$] History layer connection for dynamic bias $\hat b$
\item[$B^y$] Style layer connection for dynamic bias $\hat b$
\item[$B^h$] Hidden layer connection for dynamic bias $\hat b$
\item[$A^u_{t+1}$] Updated history layer connection for dynamic bias $\hat a$
\item[$A^v_{t+1}$] Updated visible layer connection for dynamic bias $\hat a$
\item[$A^y_{t+1}$] Updated style layer connection for dynamic bias $\hat a$
\item[$B^u_{t+1}$] Updated history layer connection for dynamic bias $\hat b$
\item[$B^y_{t+1}$] Updated style layer connection for dynamic bias $\hat b$
\item[$B^h_{t+1}$] Updated hidden layer connection for dynamic bias $\hat b$
\item[$\hat a_{t+1}$] Updated visible layer dynamic bias
\item[$\hat b_{t+1}$] Updated hidden layer dynamic bias
\item[$RelU$] Rectified linear unit activation function
\item[$\circ$] Hadamard product
\item[$RMSE$] Root mean square error
\item[$r$] Correlation coefficient
\item[$MAPE$] Mean absolute percentage error
\item[$R_t$] Real value
\item[$F_t$] Forecast value
\item[$\mu _R$] Mean of real values
\item[$\mu_F$] Mean of forecasted values\\
\end{IEEEdescription}
\end{document}
答案1
我建议你加载IEEEtrantools包。它提供了文档类的相当多的宏和环境IEEEtran
,用于其他文档类(例如类)article
。例如,IEEEtrantools
包提供了一个名为的环境IEEEdescription
,您似乎使用了它。
但是,IEEEtrantools 包不提供名为 的环境的代码IEEEkeywords
。您可能想尝试类似
\newenvironment{IEEEkeywords}{\raggedright\noindent\textsc{Keywords}:}{\par}
插入序言中。
一些额外的评论。(a)cleveref
应该加载该包最后的。(b)我认为没有任何理由在数学模式下排版 Sigmoid、RMSE 和 MAPE 这几个字。(c)我认为写 值得怀疑boltzmann
;应该写Boltzmann
。
\documentclass{article}
%% New instructions:
\usepackage{geometry} % choose suitable page parameters
\usepackage{IEEEtrantools}
\newenvironment{IEEEkeywords}{\raggedright\noindent\textsc{Keywords}:}{\par}
%% Rest of preamble may remain unchanged
\hyphenation{op-tical net-works semi-conduc-tor}
\usepackage{csquotes}
\usepackage{graphicx}
\usepackage{mathtools}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{mathtools}
\usepackage{lipsum}
\DeclareMathOperator{\RelU}{RelU}
%\usepackage{url}
%\usepackage{algorithm}
%\usepackage{array}
\usepackage{subfloat}
%\usepackage{subfig}
\usepackage{xcolor}
\usepackage{longtable}
%\usepackage{subcaption}
\usepackage{epstopdf}
\usepackage[utf8]{inputenc}
\usepackage[justification=centering]{caption}
%\usepackage{booktabs}
\newcommand{\dd}[1]{\mathrm{d}#1}
%\usepackage{amsmath}
\usepackage{graphicx}
\usepackage[utf8]{inputenc}
\usepackage[english]{babel}
\usepackage{url}
%\usepackage{multicol}
\usepackage{multirow}
\usepackage{array}
\usepackage{calc}
\usepackage[english]{babel}
%\usepackage[document]{ragged2e}
\usepackage{algorithm}
\usepackage{algpseudocode}
\usepackage{subcaption}
\usepackage{booktabs}
\usepackage{wrapfig}
\usepackage{cleveref} % this package should be loaded LAST
\title{ A deep learning approach}
\author{Ghul$^{1,2}$, Javaid$^{1*}$, \\
$^{1}$COMSATS University Islamabad 44000, Pakistan\\
$^{2}$University of Engineering and Technology, Mardan, 23200, Pakistan\\
%$^{2}$CAMS, Dept of Biomedical Technology, KSU, Riyadh 11633, Saudi Arabia\\
%$^{3}$FE, Dalhousie University, Halifax, NS B3J 4R2, Canada\\
$^{*}$Corresponding author: www.njavaid.com, [email protected]}
\usepackage{lipsum} % provides filler text
\begin{document}
\maketitle
\begin{abstract}
\lipsum*[2]
\end{abstract}
\begin{IEEEkeywords}
Short term load forecasting; deep learning; smart grid; factored conditional restricted Boltzmann machine; conditional restricted Boltzmann machine; rectified linear unit.
\end{IEEEkeywords}
\tableofcontents % optional
\section*{Nomenclature}
\addcontentsline{toc}{section}{Nomenclature}
\begin{IEEEdescription}[\IEEEusemathlabelsep%
\IEEEsetlabelwidth{Sigmoid}]
\item[$E$] {Error function}
\item[$v$] {Visible layer}
\item[$u$] {History layer}
\item[$h$] Hidden layer
\item[$a$] Visible layer bias
\item[$b$] Hidden layer bias
\item[$w^{vh}$] Bi-directional weight matrix between hidden and visible layers
\item[$w^{uh}$] Unidirectional weight matrix between hidden and visible layers
\item[$w^{uv}$] Unidirectional weight matrix between visible and history layers
\item[Sigmoid] Sigmoidal activation function
%\item[$RelU$] Rectified linear unit activation function
\item[$w^{vh}_{t+1}$] Updated bi-directional weight matrix between hidden and visible layers
\item[$w^{uh}_{t+1}$] Updated unidirectional weight matrix between hidden and visible layers
\item[$w^{uv_{t+1}}$] Updated unidirectional weight matrix between visible and history layers
\item[$a_{t+1}$] Updated bias of visible layer
\item[$b_{t+1}$] Updated bias of hidden layer
\item[$\hat a$] Visible layer dynamic bias
\item[$\hat b$] Hidden layer dynamic bias
\item[$w^v$] Visible layer weight
\item[$w^y$] Style layer weight
\item[$w^h$] Hidden layer weight
\item[$A^u$] History layer connection for dynamic bias $\hat a$
\item[$A^v$] Visible layer connection for dynamic bias $\hat a$
\item[$A^y$] Style layer connection for dynamic bias $\hat a$
\item[$B^u$] History layer connection for dynamic bias $\hat b$
\item[$B^y$] Style layer connection for dynamic bias $\hat b$
\item[$B^h$] Hidden layer connection for dynamic bias $\hat b$
\item[$A^u_{t+1}$] Updated history layer connection for dynamic bias $\hat a$
\item[$A^v_{t+1}$] Updated visible layer connection for dynamic bias $\hat a$
\item[$A^y_{t+1}$] Updated style layer connection for dynamic bias $\hat a$
\item[$B^u_{t+1}$] Updated history layer connection for dynamic bias $\hat b$
\item[$B^y_{t+1}$] Updated style layer connection for dynamic bias $\hat b$
\item[$B^h_{t+1}$] Updated hidden layer connection for dynamic bias $\hat b$
\item[$\hat a_{t+1}$] Updated visible layer dynamic bias
\item[$\hat b_{t+1}$] Updated hidden layer dynamic bias
\item[$RelU$] Rectified linear unit activation function
\item[$\circ$] Hadamard product
\item[RMSE] Root mean square error
\item[$r$] Correlation coefficient
\item[MAPE] Mean absolute percentage error
\item[$R_t$] Real value
\item[$F_t$] Forecast value
\item[$\mu _R$] Mean of real values
\item[$\mu_F$] Mean of forecasted values
\end{IEEEdescription}
\end{document}