最近，我为维基百科用 TikZ 制作了以下图片（参见文件）
我想标记绿点，就像这里所做的那样：
我尝试了几种不同的方法，但无论我做什么，文本总是在图表下方。我怎样才能用坐标标记绿点？
平均能量损失
TeX
\documentclass[varwidth=true, border=5pt]{article}
\usepackage[active,tightpage]{preview}
\usepackage[latin1]{inputenc}
\usepackage{amsmath}
\usepackage{pgfplots}
\pgfplotsset{compat=1.10}
\usepackage{tikz}
\usetikzlibrary{arrows, positioning}
\usepackage{helvet}
\usepackage[eulergreek]{sansmath}
\begin{document}
\begin{preview}
\begin{tikzpicture}
\begin{axis}[
width=13.4cm,
height=10.0cm,
% Grid
grid = major,
% size
xmin= 40, % start the diagram at this x-coordinate
xmax= 90, % end the diagram at this x-coordinate
ymin= 0, % start the diagram at this y-coordinate
ymax= 60, % end the diagram at this y-coordinate
% Legende
legend style={
font=\large\sansmath\sffamily,
at={(0.5,-0.18)},
anchor=north,
legend cell align=left,
legend columns=-1,
column sep=0.5cm
},
% Ticks
tick align=inside,
every axis/.append style={font=\large\sansmath\sffamily},
minor tick style={thick},
scaled y ticks = false,
% Axis
axis line style = {very thick,shorten <=-0.5\pgflinewidth},
axis lines = middle,
axis line style = very thick,
xlabel=Gesamtausgaben,
x label style={at={(axis description cs:0.5,-0.05)},
anchor=north,
font=\boldmath\sansmath\sffamily\Large},
ylabel=Luxusausgaben,
y label style={at={(axis description cs:-0.05,0.5)},
anchor=south,
rotate=90,
font=\boldmath\sansmath\sffamily\Large}
]
\addplot[
scatter,
only marks,
point meta=explicit symbolic,
scatter/classes={
a={mark=x,red!90!black},%
b={mark=x,cyan!50!black}},
]
table[col sep=comma, meta=label] {data.csv};
\addplot[
scatter,
only marks,
point meta=explicit symbolic,
scatter/classes={
b={mark=*,mark size=4pt,red!30!white,draw=black},%
c={mark=*,mark size=4pt,cyan!30!white,draw=black},%
a={mark=*,mark size=4pt,green!70!black,draw=black,pin=135:{\color{black}$(65, 35)$},label={(65, 35)}] {}}},
]
table[meta=label] {
x y label
65 35 a
70 40 b
60 20 c
};
% \addlegendentry{Gruppe 1}
% \addlegendentry{Gruppe 2}
\end{axis}
\end{tikzpicture}
\end{preview}
\end{document}
数据
数据由以下 Python 脚本生成
#!/usr/bin/env python
import matplotlib.pyplot as plt
import numpy
import csv
def main(n):
cov = [[25, 20], [20, 25]]
meanI = [70, 40]
datapointsI = n
meanII = [60, 20]
datapointsII = n
dataI = numpy.random.multivariate_normal(meanI, cov, datapointsI).T
x, y = dataI
plt.plot(x, y, 'x')
dataII = numpy.random.multivariate_normal(meanII, cov, datapointsII).T
x, y = dataII
plt.plot(x, y, 'x')
plt.axis('equal')
plt.show()
data = []
xs, ys = dataI
for x, y in zip(xs, ys):
data.append([x, y, 'a'])
xs, ys = dataII
for x, y in zip(xs, ys):
data.append([x, y, 'b'])
# Write data to csv files
with open("data.csv", 'wb') as csvfile:
csvfile.write("x,y,label\n")
spamwriter = csv.writer(csvfile, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)
for datapoint in data:
spamwriter.writerow(datapoint)
if __name__ == "__main__":
from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter
parser = ArgumentParser(description=__doc__,
formatter_class=ArgumentDefaultsHelpFormatter)
parser.add_argument("-n",
dest="n", default=2000, type=int,
help="how many points should get generated")
args = parser.parse_args()
main(args.n)
如果您不想执行该脚本，您也可以从存储库下载数据：https://github.com/MartinThoma/LaTeX-examples/tree/master/tikz/csv-2d-gaussian-multivarate-distributions
答案1
label和pin是node选项，但情节标记不是nodes，这就是为什么这些选项不会产生任何输出。
在这种情况下，我不建议使用\addplot命令，因为每个点都需要完全不同的样式。我只会使用普通的 TikZ 命令：
\documentclass[varwidth=true, border=5pt]{article}
\usepackage[active,tightpage]{preview}
\usepackage[latin1]{inputenc}
\usepackage{amsmath}
\usepackage{pgfplots}
\pgfplotsset{compat=1.10}
\usepackage{tikz}
\usetikzlibrary{arrows, positioning}
\usepackage{helvet}
\usepackage[eulergreek]{sansmath}
\begin{document}
\begin{preview}
\begin{tikzpicture}
\begin{axis}[
clip mode=individual,
width=13.4cm,
height=10.0cm,
% Grid
grid = major,
% size
xmin= 40, % start the diagram at this x-coordinate
xmax= 90, % end the diagram at this x-coordinate
ymin= 0, % start the diagram at this y-coordinate
ymax= 60, % end the diagram at this y-coordinate
% Legende
legend style={
font=\large\sansmath\sffamily,
at={(0.5,-0.18)},
anchor=north,
legend cell align=left,
legend columns=-1,
column sep=0.5cm
},
% Ticks
tick align=inside,
every axis/.append style={font=\large\sansmath\sffamily},
minor tick style={thick},
scaled y ticks = false,
% Axis
axis line style = {very thick,shorten <=-0.5\pgflinewidth},
axis lines = middle,
axis line style = very thick,
xlabel=Gesamtausgaben,
x label style={at={(axis description cs:0.5,-0.05)},
anchor=north,
font=\boldmath\sansmath\sffamily\Large},
ylabel=Luxusausgaben,
y label style={at={(axis description cs:-0.05,0.5)},
anchor=south,
rotate=90,
font=\boldmath\sansmath\sffamily\Large}
]
\addplot[
scatter,
only marks,
point meta=explicit symbolic,
scatter/classes={
a={mark=*,red!90!black, mark size=1, opacity=0.3, on layer=axis background},%
b={mark=*,cyan!50!black, mark size=1, opacity=0.3}},
]
table[col sep=comma, meta=label] {data.csv};
\filldraw [fill=red!50, draw=white, thick] (axis cs:70,40) circle [radius=4pt];
\filldraw [fill=cyan!50, draw=white, thick] (axis cs:60,20) circle [radius=4pt];
\filldraw [fill=yellow, draw=black, thick] (axis cs:65,35) circle [radius=4pt] node [label={[inner sep=1pt, fill=white,text=black, fill opacity=0.75, text opacity=1]above left:$(65, 35)$}] {};
% \addlegendentry{Gruppe 1}
% \addlegendentry{Gruppe 2}
\end{axis}
\end{tikzpicture}
\end{preview}
\end{document}