正如我所说，我正在为学校编写数学书；其中一本是关于矩阵的，我对它遇到了几个问题。
这是一个很大的代码，所以我尝试总结一下：
TikZ 矩阵的问题：当我尝试编译它时
\draw [<->, yl] (A-2-4.north) to [bend left] node [arrow style mul] {$+$} (B-2-4.north);使用该
tikz包和以下tikz库：\usepackage{tikz} \usetikzlibrary{decorations.pathmorphing} \usetikzlibrary{decorations.markings} \usetikzlibrary{shapes,snakes,calc,arrows} \usetikzlibrary{intersections,through,backgrounds} \usetikzlibrary{shapes.arrows,chains,positioning} \usetikzlibrary{matrix,decorations.pathmorphing}我明白了
No shaped name A-2-4 is known. See the pgf package documentation for explanation当我尝试编译它时（使用
\include{filename}）：\chapterimage{m2summary.pdf} \chapter{Respostas e soluções} \begin{multicols}{2} \section*{\textcolor{ocre}{1. Matrizes}} \subsection*{Exercícios de fixação} \begin{enumerate}[label=\arabic*.] \tiny{ \item $\ds A=\begin{bmatrix} 0 & -1 & -2 \\ 1 & 0 & -1 \\ 2 & 1 & 0 \end{bmatrix}$ } \end{enumerate} \end{multicols}表明：
Missing $ inserted。怎么办？我尝试在没有文本的情况下进行编译，实际上，没有数学文本，但它一直显示缺少 $。我真的不知道该怎么办！
抱歉写了这么长的帖子，我只是尽量写得完整。我使用的是 WinEdt 8，MikTeX 2.9，是我这周（2014 年 2 月）安装的。
TIKZ 图片代码：
\begin{tikzpicture}
\matrix(A)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (0,0){
a_{11} & a_{12} & \cdots & a_{1n} \\
\node[node style sp] {a_{21}}; & \node[node style sp] {a_{22}}; & \cdots & \node[node style sp] {a_{2n}}; \\
\vdots & \vdots & \ddots & \vdots \\
a_{m1} & a_{m2} & \cdots & a_{mn} \\
};
\node [draw, above=10pt] at (A.north) {matriz $\color{ocre}A$};
\node [right=20pt] at (A.east) {$\color{ocre}+$};
\matrix(B)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (6,0){
b_{11} & b_{12} & \cdots & b_{1n} \\
\node[node style sp] {b_{21}}; & \node[node style sp] {b_{22}}; & \cdots & \node[node style sp] {b_{2n}}; \\
\vdots & \vdots & \ddots & \vdots \\
b_{m1} & b_{m2} & \cdots & b_{mn} \\
};
\node [draw, above=10pt] at (B.north) {matriz $\color{ocre}B$};
\matrix(C)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (6,-4.5){
c_{11} & c_{12} & \cdots & c_{1n} \\
\node[node style sp, ocre] {c_{21}}; & \node[node style sp, blue] {c_{22}}; & \cdots & \node[node style sp, yl] {c_{2n}}; \\
\vdots & \vdots & \ddots & \vdots \\
c_{m1} & c_{m2} & \cdots & c_{mn} \\
};
\node [draw, below=10pt] at (C.south) {matriz $\color{ocre}C=A+B$};
\node (a) [above=10pt] at (A-2-1.north) {};
\node (b) [above=10pt] at (C-2-1.north) {};
\draw [<->, yl] (A-2-4.north) to [bend left] node [arrow style mul] {$+$} (B-2-4.north);
\draw [->, yl] (B-2-4.south) to (C-2-4.north);
\draw [<->, blue] (A-2-2.north) to [bend left] node [arrow style mul] {$+$} (B-2-2.north);
\draw [->, blue] (B-2-2.south) to (C-2-2.north);
\draw [<->, ocre] (A-2-1.north) to [bend left] node [arrow style mul] {$+$} (B-2-1.north);
\draw [->, ocre] (B-2-1.south) to (C-2-1.north);
\end{tikzpicture}
答案1
（这只回答了 TikZ 问题。对于其他问题，我建议将代码简化为最小工作示例（MWE）并就此提出一个新问题。）
您的问题是您有一个matrix of math nodes，但您\node {}在单元格中使用。默认情况下，当您有一个matrix of( math)时，TikZ 会在单元格内容前面nodes添加类似内容（并在其后添加结束符），但\node [name=A-1-2] {};不是当你\node在单元格中使用或类似内容时。引用手册：
从概念上讲，此键
\node{在每个单元格的开头和};结尾添加，并将anchor节点的设置为base。此外，它还name为每个节点添加选项 option，其中名称设置为 矩阵名称-行号-列号。例如，如果矩阵的名称为my matrix，则左上角单元格中的节点将命名为 mymatrix-1-1。[...]
如果您的单元格以
\path命令或任何扩展为 的命令开头\path，其中包括\draw、\node和\fill其他命令，则\node{启动代码和};代码将被抑制。
为了能够将选项/样式传递给节点，请写入例如
|[my node style]| content of matrix cell
即在包含选项的括号周围添加竖线，并将其放在单元格的开头。
下面是完整代码。您没有添加样式定义，因此我将它们定义为空样式。
\documentclass[11pt]{standalone}
\usepackage{tikz}
\usetikzlibrary{matrix}
\tikzset{
node style ge/.style={},
node style sp/.style={},
yl/.style={},
arrow style mul/.style={},
}
\definecolor{ocre}{RGB}{204, 119, 34}
\begin{document}
\begin{tikzpicture}
\matrix(A)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (0,0){
a_{11} & a_{12} & \cdots & a_{1n} \\
| [node style sp]| a_{21} & |[node style sp]| a_{22} & \cdots & |[node style sp]| a_{2n} \\
\vdots & \vdots & \ddots & \vdots \\
a_{m1} & a_{m2} & \cdots & a_{mn} \\
};
\node [draw, above=10pt,font=\color{ocre}] at (A.north) {matriz $A$};
\node [right=20pt,font=\color{ocre}] at (A.east) {$+$};
\matrix(B)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (6,0){
b_{11} & b_{12} & \cdots & b_{1n} \\
|[node style sp]| b_{21} & |[node style sp]| b_{22} & \cdots & |[node style sp]| b_{2n} \\
\vdots & \vdots & \ddots & \vdots \\
b_{m1} & b_{m2} & \cdots & b_{mn} \\
};
\node [draw, above=10pt,font=\color{ocre}] at (B.north) {matriz $B$};
\matrix(C)[matrix of math nodes, nodes={node style ge}, left delimiter={[}, right delimiter={]}] at (6,-4.5){
c_{11} & c_{12} & \cdots & c_{1n} \\
|[node style sp,ocre]| c_{21} & |[node style sp, blue]| c_{22} & \cdots & |[node style sp, yl]| c_{2n} \\
\vdots & \vdots & \ddots & \vdots \\
c_{m1} & c_{m2} & \cdots & c_{mn} \\
};
\node [draw, below=10pt,font=\color{ocre}] at (C.south) {matriz $C=A+B$};
\node (a) [above=10pt] at (A-2-1.north) {};
\node (b) [above=10pt] at (C-2-1.north) {};
\draw [<->, yl] (A-2-4.north) to [bend left] node [arrow style mul] {$+$} (B-2-4.north);
\draw [->, yl] (B-2-4.south) to (C-2-4.north);
\draw [<->, blue] (A-2-2.north) to [bend left] node [arrow style mul] {$+$} (B-2-2.north);
\draw [->, blue] (B-2-2.south) to (C-2-2.north);
\draw [<->] (A-2-1.north) to [bend left] node [arrow style mul] {$+$} (B-2-1.north);
\draw [->] (B-2-1.south) to (C-2-1.north);
\end{tikzpicture}
\end{document}