我从中学到了一种制作神经网络(nn)的好方法这个答案
当我为我的个人 nn 编辑它时,如下所示:
\documentclass[article]{standalone}
\usepackage{tikz}
\usepackage{etoolbox} % for \ifnumcomp
\usepackage{listofitems} % for \readlist to create arrays
\tikzset{>=latex} % for LaTeX arrow head
\colorlet{myred}{red!80!black}
\colorlet{myblue}{blue!80!black}
\colorlet{mygreen}{green!60!black}
\colorlet{myorange}{orange!70!red!60!black}
\colorlet{mydarkred}{red!30!black}
\colorlet{mydarkblue}{blue!40!black}
\colorlet{mydarkgreen}{green!30!black}
\definecolor{iceberg}{rgb}{0.44, 0.65, 0.82}
\definecolor{firebrick}{rgb}{0.7, 0.13, 0.13}
\definecolor{outerspace}{rgb}{0.25, 0.29, 0.3}
\definecolor{lemonchiffon}{rgb}{1.0, 0.98, 0.8}
\definecolor{flavescent}{rgb}{0.97, 0.91, 0.56}
\definecolor{floralwhite}{rgb}{1.0, 0.98, 0.94}
\definecolor{airforceblue}{rgb}{0.36, 0.54, 0.66}
\definecolor{electriclime}{rgb}{0.8, 1.0, 0.0}
\tikzstyle{node}=[thick,circle,draw=myblue,minimum size=22,inner sep=0.5,outer sep=0.6]
\tikzstyle{node in}=[node,myblue!30!black,draw=myblue!30!black,fill=floralwhite]
\tikzstyle{node hidden}=[node,blue!20!black,draw=myblue!5!black,fill=airforceblue!15]
\tikzstyle{node convol}=[node,orange!20!black,draw=myorange!30!black,fill=myorange!20]
\tikzstyle{node out}=[node,red!20!black,draw=myred!30!black,fill=firebrick!20]
\tikzstyle{connect}=[thick,mydarkblue] %,line cap=round
\tikzstyle{connect arrow}=[-{Latex[length=4,width=3.5]},thick,mydarkblue,shorten <=0.5,shorten >=1]
\tikzset{ % node styles, numbered for easy mapping with \nstyle
node 1/.style={node in},
node 2/.style={node hidden},
node 3/.style={node out},
}
\def\nstyle{int(\lay<\Nnodlen?min(2,\lay):3)} % map layer number onto 1, 2, or 3
\begin{document}
% NEURAL NETWORK
\begin{tikzpicture}[x=2.2cm,y=1.4cm]
\message{^^JNeural network, shifted}
\readlist\Nnod{3,5,5,5,5,5,1} % array of number of nodes per layer
\readlist\Nstr{n,m,m,m,k} % array of string number of nodes per layer
\readlist\Cstr{\strut x,a^{(\prev)},a^{(\prev)},a^{(\prev)},y} % array of coefficient symbol per layer
\def\yshift{0.5} % shift last node for dots
\message{^^J Layer}
\foreachitem \N \in \Nnod{ % loop over layers
\def\lay{\Ncnt} % alias of index of current layer
\pgfmathsetmacro\prev{int(\Ncnt-1)} % number of previous layer
\message{\lay,}
\foreach \i [evaluate={\c=int(\i==\N); \y=\N/2-\i-\c*\yshift;
\index=(\i<\N?int(\i):"\Nstr[\lay]");
\x=\lay; \n=\nstyle;}] in {1,...,\N}{ % loop over nodes
% NODES
\node[node \n] (N\lay-\i) at (\x,\y){}; %{$\Cstr[\lay]_{\index}$};
% CONNECTIONS
\ifnum\lay>1 % connect to previous layer
\foreach \j in {1,...,\Nnod[\prev]}{ % loop over nodes in previous layer
\draw[connect,white,line width=1.2] (N\prev-\j) -- (N\lay-\i);
\draw[connect] (N\prev-\j) -- (N\lay-\i);
%\draw[connect] (N\prev-\j.0) -- (N\lay-\i.180); % connect to left
}
\fi % else: nothing to connect first layer
}
\path (N\lay-\N) --++ (0,1+\yshift) node[midway,scale=1.5] {$\vdots$};
}
% LABELS
\node[above=0.8,align=center,myblue!50!black] at (N1-1.90) {Input\\[-0.2em]Layers};
\node[above=0,align=center,airforceblue] at (N4-1.90) {Hidden Layers};
\node[above=60,align=center,myred!60!black] at (N\Nnodlen-1.90) {Output\\[-0.2em]Layer};
\end{tikzpicture}
\end{document}
如您所见,有三个垂直点($\vdots$
),它们表示“输入层”、“隐藏层”和“输出层”之间的节点迭代。当我尝试仅删除(或取消)$\vdots$
“输出层”时,所有点$\vdots$
都消失了。所以我必须保留$\vdots$
如下内容:
我的问题是:我怎样才能只保留$\vdots$
输入层和隐藏层,而不保留输出层?或者如何设置任何单个层应$\vdots$
相应地排除迭代符号?
提前谢谢您,我希望附图能帮助我解释我的问题。
答案1
概括\tikzCenterNodes
宏可以重新创建\tikzMatrixNodes
此图,而无需...代码中发生的一切。
此外,该graphs
库还帮助绘制所有线条。
有一种方法parse
可以循环遍历所有隐藏层,但我想不出一个干净的方法。在这里,我将手动添加节点,但使用循环\foreach
跳过\vdots
节点。
代码
\documentclass[tikz]{standalone}
\usetikzlibrary{ext.positioning-plus,graphs}
\tikzset{
node matrix/.style={row sep=y_node_dist, column sep=x_node_dist,
every outer matrix/.append style={/pgf/inner sep=+0pt, /pgf/outer sep=+0pt, draw=none, fill=none, shape=rectangle}},
node matrix/node/.style={node contents=,anchor=center,
name/.expanded={\tikzmatrixname_\ifnum\pgfmatrixcurrentrow=1 \the\pgfmatrixcurrentcolumn\else\the\pgfmatrixcurrentrow\fi},
{nm \ifnum\pgfmatrixcurrentrow=1 \the\pgfmatrixcurrentcolumn\else\the\pgfmatrixcurrentrow\fi}/.try},
node matrix/place 1st node/.code args={#1,#2}{\node[node matrix/node,#1];},
node matrix/place other nodes/.style args={#1,#2}{/tikz/node matrix/place oth node/.list={#2}},
vertical node matrix/.style={/tikz/node matrix/place oth node/.code={\pgfmatrixendrow \node[node matrix/node,##1];}},
horizontal node matrix/.style={/tikz/node matrix/place oth node/.code={\pgfmatrixnextcell\node[node matrix/node,##1];}}}
\newcommand*\tikzMatrixNodes[2][1]{%
\matrix[every node matrix/.try,node matrix,#1]{
\tikzset{node matrix/place 1st node={#2},node matrix/place other nodes={#2}}\\};}
\begin{document}
\begin{tikzpicture}[
node distance=5mm and 10mm,
layers/.style={circle, thick, draw, minimum size=+5mm},
input/.style ={layers, fill=yellow!30},
hidden/.style={layers, fill=blue!30},
output/.style={layers, fill=red!30},
every node matrix/.style={vertical node matrix},
second to last/.style={
nm #1/.style={node contents=\vdots, text height=2ex, fill=none, draw=none, minimum size=+0pt,inner sep=+0pt, shape=rectangle},
row #1/.append style={row sep=.5*y_node_dist}, row \inteval{#1-1}/.append style={row sep=.5*y_node_dist}},
]
\tikzMatrixNodes[name=hidden0, nodes=input, second to last=3]{,,,}
\foreach \hidden[count=\lastHidden from 0] in {1,...,5}{% chains?
\tikzMatrixNodes[
right=of hidden\lastHidden, name=hidden\hidden,
nodes=hidden, second to last=5]{,,,,,}}
\node[right=of hidden5, output](output){};
\path[thick, blue] graph[use existing nodes]{
{\foreach \x in {1,2,4}{hidden0_\x}}
-- [complete bipartite] {\foreach \x in {1,...,4,6}{hidden1_\x}}
-- [complete bipartite] {\foreach \x in {1,...,4,6}{hidden2_\x}}
-- [complete bipartite] {\foreach \x in {1,...,4,6}{hidden3_\x}}
-- [complete bipartite] {\foreach \x in {1,...,4,6}{hidden4_\x}}
-- [complete bipartite] {\foreach \x in {1,...,4,6}{hidden5_\x}}
-- {output};
};
\path[nodes={align=center}]
(hidden3_1.north) node[above,blue!70] (hl) {Hidden Layers} % could've been a label
(hidden0_1|-hl.south) node[above,blue] {Input\\Layers}
(output|-hl.south) node[above,red] {Output\\Layers};
\end{tikzpicture}
\end{document}
输出
答案2
我还没有尝试去理解代码——它一点也不简单。我只是注意到它使用了\ifnum\lay ... \fi
——所以你可以这样做:
\documentclass[tikz, border=1cm]{standalone}
\usepackage{etoolbox} % for \ifnumcomp
\usepackage{listofitems} % for \readlist to create arrays
\tikzset{>=latex} % for LaTeX arrow head
\colorlet{myred}{red!80!black}
\colorlet{myblue}{blue!80!black}
\colorlet{mygreen}{green!60!black}
\colorlet{myorange}{orange!70!red!60!black}
\colorlet{mydarkred}{red!30!black}
\colorlet{mydarkblue}{blue!40!black}
\colorlet{mydarkgreen}{green!30!black}
\definecolor{iceberg}{rgb}{0.44, 0.65, 0.82}
\definecolor{firebrick}{rgb}{0.7, 0.13, 0.13}
\definecolor{outerspace}{rgb}{0.25, 0.29, 0.3}
\definecolor{lemonchiffon}{rgb}{1.0, 0.98, 0.8}
\definecolor{flavescent}{rgb}{0.97, 0.91, 0.56}
\definecolor{floralwhite}{rgb}{1.0, 0.98, 0.94}
\definecolor{airforceblue}{rgb}{0.36, 0.54, 0.66}
\definecolor{electriclime}{rgb}{0.8, 1.0, 0.0}
\tikzstyle{node}=[thick,circle,draw=myblue,minimum size=22,inner sep=0.5,outer sep=0.6]
\tikzstyle{node in}=[node,myblue!30!black,draw=myblue!30!black,fill=floralwhite]
\tikzstyle{node hidden}=[node,blue!20!black,draw=myblue!5!black,fill=airforceblue!15]
\tikzstyle{node convol}=[node,orange!20!black,draw=myorange!30!black,fill=myorange!20]
\tikzstyle{node out}=[node,red!20!black,draw=myred!30!black,fill=firebrick!20]
\tikzstyle{connect}=[thick,mydarkblue] %,line cap=round
\tikzstyle{connect arrow}=[-{Latex[length=4,width=3.5]},thick,mydarkblue,shorten <=0.5,shorten >=1]
\tikzset{ % node styles, numbered for easy mapping with \nstyle
node 1/.style={node in},
node 2/.style={node hidden},
node 3/.style={node out},
}
\def\nstyle{int(\lay<\Nnodlen?min(2,\lay):3)} % map layer number onto 1, 2, or 3
\begin{document}
% NEURAL NETWORK
\begin{tikzpicture}[x=2.2cm,y=1.4cm]
\message{^^JNeural network, shifted}
\readlist\Nnod{3,5,5,5,5,5,1} % array of number of nodes per layer
\readlist\Nstr{n,m,m,m,k} % array of string number of nodes per layer
\readlist\Cstr{\strut x,a^{(\prev)},a^{(\prev)},a^{(\prev)},y} % array of coefficient symbol per layer
\def\yshift{0.5} % shift last node for dots
\message{^^J Layer}
\foreachitem \N \in \Nnod{ % loop over layers
\def\lay{\Ncnt} % alias of index of current layer
\pgfmathsetmacro\prev{int(\Ncnt-1)} % number of previous layer
\message{\lay,}
\foreach \i [evaluate={\c=int(\i==\N); \y=\N/2-\i-\c*\yshift;
\index=(\i<\N?int(\i):"\Nstr[\lay]");
\x=\lay; \n=\nstyle;}] in {1,...,\N}{ % loop over nodes
% NODES
\node[node \n] (N\lay-\i) at (\x,\y){}; %{$\Cstr[\lay]_{\index}$};
% CONNECTIONS
\ifnum\lay>1 % connect to previous layer
\foreach \j in {1,...,\Nnod[\prev]}{ % loop over nodes in previous layer
\draw[connect,white,line width=1.2] (N\prev-\j) -- (N\lay-\i);
\draw[connect] (N\prev-\j) -- (N\lay-\i);
%\draw[connect] (N\prev-\j.0) -- (N\lay-\i.180); % connect to left
}
\fi % else: nothing to connect first layer
}
\ifnum\lay<7 \path (N\lay-\N) --++ (0,1+\yshift) node[midway,scale=1.5] {$\vdots$}; \fi
}
% LABELS
\node[above=0.8,align=center,myblue!50!black] at (N1-1.90) {Input\\[-0.2em]Layers};
\node[above=0,align=center,airforceblue] at (N4-1.90) {Hidden Layers};
\node[above=60,align=center,myred!60!black] at (N\Nnodlen-1.90) {Output\\[-0.2em]Layer};
\end{tikzpicture}
\end{document}