我有一些由 matlab2tikz 生成的图，我想将其叠加。每个图都是一个单独的 .tikz 文件，我使用 \input 输入。我的代码目前是：
\pgfplotsset{
every axis/.append style={
scale only axis,
},
/tikz/every picture/.append style={
trim axis left,
trim axis right,
}
}
\newcolumntype{P}[1]{>{\centering\arraybackslash}p{#1}}
\begin{table}[h!]\centering
\begin{center}
\begin{tabular}{P{6cm} P{6cm}}
\toprule
R220expt21 & R60expt\\
\cmidrule(r){1-1}\cmidrule(l){2-2}
\scalebox{0.4}{\input{images/expt21nrmse.tikz}}\llap{\raisebox{0.5cm}{\scalebox{0.2} {\input{images/expt21nrmse.tikz}}}}
\scalebox{0.4}{\input{images/expt21fitD.tikz}}
&
\scalebox{0.4}{\input{images/expt2nrmse.tikz}}
\scalebox{0.4}{\input{images/expt2fitD.tikz}}
\\ \bottomrule
\end{tabular}
\caption{Fitted results for two samples. (top) NRMSE values for each q. (bottom) Fitted diffusion. (bottom insert) 2 Highlighted DICF and theoretical fits.}
\label{tbl:1}
\end{center}
\end{table}
目前此代码产生：
如您所见，左上角图像的插图看起来不错，但我想将其水平移动。我尝试了主题“水平等同于 \raisebox？”中的解决方案，但出现有关处于水平模式的错误。
我的问题是：
- 有没有一种方法，与我目前使用 \input 所做的一样，可以整齐地定位子图？
- 我的方法是否不好？如果是，我该怎么办？
谢谢你的时间。
编辑2：完整MWE：
\documentclass[11pt,a4paper]{report}
\usepackage[a4paper,vmargin={30mm,30mm},hmargin={30mm,30mm}]{geometry}
\usepackage{graphicx,array,booktabs,pgfplots}
\usetikzlibrary{calc}
\newsavebox{\mybox}
%\sbox{\mybox}{\input{images/exp5nrmse.tikz}}
\sbox{\mybox}{% Just use \input{images/exp5nrmse.tikz} instead of following
\begin{tikzpicture}
\begin{axis}[%
width=4.52083333333333in,
height=3.565625in,
scale only axis,
xmin=1,
xmax=511,
xlabel={$q (\mu m^{-1})$},
ymin=0.5,
ymax=1,
ylabel={Normalized RMS error}
]
\addplot [color=blue,solid,line width=1.0pt,forget plot]
table[row sep=crcr]{
10 0.49608542 \\
11 0.5509195 \\
12 0.54464869 \\
13 0.54379236 \\
14 0.60808362 \\
15 0.49726539 \\
16 0.61919337 \\
17 0.66339954 \\
18 0.72671426 \\
19 0.71607429 \\
};
\end{axis}
\end{tikzpicture}%
}
\pgfplotsset{
every axis/.append style={
scale only axis,
},
/tikz/every picture/.append style={
trim axis left,
trim axis right,
}
}
\begin{document}
\newcolumntype{P}[1]{>{\centering\arraybackslash}p{#1}}
\begin{table}[htb!]
\centering
\begin{tabular}{P{5.5cm} P{5.5cm}}
\toprule
R220expt21 & R60expt\\
\cmidrule(r){1-1}\cmidrule(l){2-2}
\tikz[remember picture,baseline=(a.base)]\node[scale=0.4,outer sep=0pt]
(a){\usebox{\mybox}};
%%% Change dimensions in $(a.east)+(-0.9,0cm)$
\tikz[remember picture,overlay]\node[anchor=east,outer sep=0pt,scale=0.2] at
($(a.east)+(-0.9,0cm)$){\usebox{\mybox}};
\scalebox{0.4}{\usebox{\mybox}}
&
\scalebox{0.4}{\usebox{\mybox}}
\scalebox{0.4}{\usebox{\mybox}}
\\ \bottomrule
\end{tabular}
\caption{Fitted results for two samples. (top) NRMSE values for each q. (bottom) Fitted diffusion. (bottom insert) 2 Highlighted DICF and theoretical fits.}
\label{tbl:1}
\end{table}
\end{document}
答案1
希望这也能起作用\input。
您可以使用来node放置您的.tikz文件
\documentclass{article}
\usepackage{graphicx,array,booktabs,tikz}
\usetikzlibrary{calc}
\begin{document}
\newcolumntype{P}[1]{>{\centering\arraybackslash}p{#1}}
\begin{table}[htb!]
\centering
\begin{tabular}{P{5.5cm} P{5.5cm}}
\toprule
R220expt21 & R60expt\\
\cmidrule(r){1-1}\cmidrule(l){2-2}
\tikz[remember picture,baseline=(a.base)]\node[scale=0.4,outer sep=0pt]
(a){\includegraphics[width=6cm]{example-image-a}};
%%% Change dimensions in $(a.east)+(-0.5,0cm)$
\tikz[remember picture,overlay]\node[anchor=east,outer sep=0pt,scale=0.2] at
($(a.east)+(-0.5,0cm)$){\includegraphics[width=6cm]{example-image-b}};
\scalebox{0.4}{\includegraphics[width=6cm]{example-image-c}}
&
\scalebox{0.4}{\includegraphics[width=6cm]{example-image-a}}
\scalebox{0.4}{\includegraphics[width=6cm]{example-image}}
\\ \bottomrule
\end{tabular}
\caption{Fitted results for two samples. (top) NRMSE values for each q. (bottom) Fitted diffusion. (bottom insert) 2 Highlighted DICF and theoretical fits.}
\label{tbl:1}
\end{table}
\end{document}
更新：
\documentclass{article}
\usepackage{graphicx,array,booktabs,pgfplots}
\pgfplotsset{compat=1.9}
\usetikzlibrary{calc}
%% use as many boxes as you need for each plot.
\newsavebox{\mybox}
%\sbox{\mybox}{\input{images/exp5nrmse.tikz}}
\sbox{\mybox}{% Just use \input{images/exp5nrmse.tikz} instead of following
\begin{tikzpicture}
\begin{axis}[%
width=4.52083333333333in,
height=3.565625in,
scale only axis,
xmin=1,
xmax=511,
trim axis left,
trim axis right,
xlabel={$q (\mu m^{-1})$},
ymin=0.5,
ymax=1,
ylabel={Normalized RMS error}
]
\addplot [color=blue,solid,line width=1.0pt,forget plot]
table[row sep=crcr]{
10 0.49608542 \\
11 0.5509195 \\
12 0.54464869 \\
13 0.54379236 \\
14 0.60808362 \\
15 0.49726539 \\
16 0.61919337 \\
17 0.66339954 \\
18 0.72671426 \\
19 0.71607429 \\
};
\end{axis}
\end{tikzpicture}%
}%
\begin{document}
\newcolumntype{P}[1]{>{\centering\arraybackslash}p{#1}}
\begin{table}[htb!]
\centering
\begin{tabular}{P{5.5cm} P{5.5cm}}
\toprule
R220expt21 & R60expt\\
\cmidrule(r){1-1}\cmidrule(l){2-2}
\begin{tikzpicture}[baseline=(a.base)]
\node[scale=0.4,outer sep=0pt] (a){\usebox{\mybox}};
%%% Change dimensions in $(a.east)+(-0.9,0cm)$
\node[anchor=east,outer sep=0pt,scale=0.2] at
($(a.east)+(-0.9,0cm)$){\usebox{\mybox}};
\end{tikzpicture}
\scalebox{0.4}{\usebox{\mybox}}
&
\scalebox{0.4}{\usebox{\mybox}}
\scalebox{0.4}{\usebox{\mybox}}
\\ \bottomrule
\end{tabular}
\caption{Fitted results for two samples. (top) NRMSE values for each q. (bottom) Fitted diffusion. (bottom insert) 2 Highlighted DICF and theoretical fits.}
\label{tbl:1}
\end{table}
\end{document}
