[Tex/LaTex] How to properly arrange group plot for bar chart using pgfplot

groupplotspgfplotstikz-pgf

I'm using pgfplot for plotting in Latex. I want to achieve this kind of graph

And this is the best I can do so far [Update #4],

Here's my latex code [Updated #4] :

\begin{tikzpicture}
\begin{groupplot}[
       group name=plot,
       group size=2 by 1,
       xlabels at=edge bottom,
       ylabels at=edge left,
       horizontal sep=-10pt,
       vertical sep=0pt,
      /pgf/bar width=12pt},
legend columns=-1,
legend  style={at={(6.5 ,3.6)}},
legend entries={{\color{black}{\footnotesize Win}},{\color{black}{\footnotesize Lose}}},
legend to name=CombinedLegendBar,
ybar legend,
ylabel={Participant},
major x tick style=transparent,
ybar= 1*\pgflinewidth,
ymax=6,
width=6cm,
x axis line style={opacity=0},
x tick label style={rotate=0, anchor=center},
symbolic x coords={{Task 1}, {Task 2}, {Task 3}},
xticklabel style={yshift=-2mm,xshift={ifthenelse(\ticknum==2,0,0)}}, 
xtick=data,
ymajorgrids=true,
grid style=dashed,
nodes near coords,
scale only axis,
point meta=explicit symbolic
]

%modifier
\nextgroupplot[
     xlabel=Multiplayer,
     width=0.2\textwidth,
     height=0.2\textwidth,
     enlarge x limits = 0.5]
\addplot[draw=green,thick,fill=green,fill opacity=0.6] coordinates {
  (Task 1,4)[\textcolor{blue!60}{4}]
  (Task 2,4)[\textcolor{blue!60}{4}]
  (Task 3,4)[\textcolor{blue!60}{4}]};

\addplot[draw=orange,thick,fill=orange,fill opacity=0.6] coordinates {
  (Task 1,2)[\textcolor{orange!100}{2}]
  (Task 2,1)[\textcolor{orange!100}{1}]
  (Task 3,3)[\textcolor{orange!100}{3}]};
  
\nextgroupplot[
     xlabel=Single Player,
     yticklabels={,,},
     width=0.13\textwidth,
     height=0.2\textwidth,
     enlarge x limits = 0.5]
\addplot[draw=green,thick,fill=green,fill opacity=0.6] coordinates {
  (Task 1,4)[\textcolor{blue!60}{4}]
  (Task 2,4)[\textcolor{blue!60}{4}]};

\addplot[draw=orange,thick,fill=orange,fill opacity=0.6] coordinates {
  (Task 1,2)[\textcolor{orange!100}{2}]
  (Task 2,1)[\textcolor{orange!100}{1}]};
\end{groupplot}

\end{tikzpicture}
\ref{CombinedLegendBar}

My problem now :

~~My plot looks untidy, the space between each groupplot also doesn't look nice~~.
My solution : set width, height and enlarge x limits for EACH group plot (\nextgroupplot) instead of on global setting (group plot setting). Don't forget "scale only axis" to help modify width and height properly.
The dashed line overlapped between two group plot. So what I notice is at the tick of the grid is solid at the end and beginning, that's why when it overlaps with another dashed line, it doesn't look nice.
Graph title. I tried to use \node to put my title (of all subplots). Doesn't work properly.
~~Is it possible to put my legend on the upper empty space inside of the graph ?~~ DONE !! Use legend setting and call it with \ref

Any advice ?

Best Answer

You could perhaps do some calculations to get useful widths, see example below. Not sure they're "correct", but the result looked OK.
ytick pos=left for the first groupplot, ytick pos=right for the second.
No idea what you mean by "doesn't work properly", adding a node at a useful location works fine. But of course, if you want a numbered caption, use \caption.
Do you really need legend to name and all that? Seems like just adding legend columns=-1 is enough (given that the height of the diagram is sufficient).

There are also some other modifications, such as making a custom cycle list, and setting the style of the nodes near coords instead of using \textcolor everywhere.

That said, it is also possible to do this with just one axis, see second tikzpicture in code below.

\documentclass[tikz,border=5mm]{standalone}
\usepackage{pgfplots}
\usepgfplotslibrary{groupplots}
\pgfplotsset{compat=1.16}
\begin{document}
\begin{tikzpicture}

\pgfmathsetmacro{\NPlotsFirstAx}{3}
\pgfmathsetmacro{\NPlotsSecondAx}{2}
\pgfmathsetmacro{\SumPlots}{\NPlotsFirstAx+\NPlotsSecondAx+2}
\pgfmathsetmacro{\WdFirst}{\NPlotsFirstAx/\SumPlots}
\pgfmathsetmacro{\WdSecond}{\NPlotsSecondAx/\SumPlots}
\begin{groupplot}[
       group style={
       group name=plot,
       group size=2 by 1,
       xlabels at=edge bottom,
       ylabels at=edge left,
       horizontal sep=0pt,
       vertical sep=0pt,
       /pgf/bar width=10pt},
ylabel={Participant},
major x tick style=transparent,
ybar= \pgflinewidth,
ymax=6,
%width=6cm,
x axis line style={opacity=0},
x tick label style={rotate=0, anchor=center},
%symbolic x coords={{Task 1}, {Task 2}, {Task 3}},
xticklabel style={yshift=-2mm,xshift={ifthenelse(\ticknum==2,0,0)}}, 
xtick=data,
xticklabels={Task 1, Task 2, Task 3},
ymajorgrids=true,
grid style=dotted,
nodes near coords,
scale only axis,
point meta=explicit symbolic,
enlarge x limits = {abs=1},
cycle list={
  draw=blue,thick,fill=blue,fill opacity=0.6,nodes near coords style={blue!60}\\
  draw=orange,thick,fill=orange,fill opacity=0.6,nodes near coords style={orange}\\
  },
legend columns=-1,
height=0.3\textwidth,
]

%modifier
\nextgroupplot[
     xlabel=Color,
     width=\WdFirst\textwidth,
     ytick pos=left
      ]
\addplot  coordinates {
  (1,4)[4]
  (2,4)[4]
  (3,4)[4]};

\addplot  coordinates {
  (1,2)[2]
  (2,1)[1]
  (3,3)[3]};

\nextgroupplot[
     xlabel=Brightness,
     yticklabels={},
     width=\WdSecond\textwidth,
     ytick pos=right
     ]
\addplot coordinates {
  (1,4)[4]
  (2,4)[4]};

\addplot coordinates {
  (1,2)[2]
  (2,1)[1]};

\legend{Win, Lose}
\end{groupplot}


\node [above] at (current bounding box.north) {Title of entire thing};

\end{tikzpicture}

\begin{tikzpicture}
\begin{axis}[
  xticklabels={Task 1, Task 2, Task 3, Task 1, Task 2},
  xtick=data,
  ylabel={Participant},
  major x tick style=transparent,
  ybar= \pgflinewidth,
  ymax=6,
  x axis line style={opacity=0},
  x tick label style={rotate=0, anchor=center},
  xticklabel style={yshift=-2mm}, 
  xtick=data,
  ymajorgrids=true,
  grid style=dotted,
  nodes near coords,
  scale only axis,
  point meta=explicit symbolic,
  enlarge x limits = {abs=1},
  cycle list={
    draw=blue,thick,fill=blue,fill opacity=0.6,nodes near coords style={blue!60}\\
    draw=orange,thick,fill=orange,fill opacity=0.6,nodes near coords style={orange}\\
  },
  legend columns=-1,
  legend pos=north east,
  height=0.3\textwidth,
  width=0.7\textwidth,
  extra x ticks={2, 5.5},
  extra x tick labels={Color,Brightness},
  extra x tick style={yshift=-15pt},
  title={Title of entire thing}
]

\addplot  coordinates {
  (1,4)[4]
  (2,4)[4]
  (3,4)[4]
  (5,4)[4]
  (6,4)[4]};


\addplot  coordinates {
  (1,2)[2]
  (2,1)[1]
  (3,3)[3]
  (5,2)[2]
  (6,1)[1]};

\legend{Win,Lose}
\end{axis}
\end{tikzpicture}
\end{document}

Code 1

\begin{axis}
[   ybar,
  xlabel=Character,
  ylabel=Probability,
  xtick=data,
  xticklabels from table={my.dat}{Character},
  bar width=2pt,
  width=0.95*\textwidth,
  x tick label style={font=\tiny,align=right,rotate=90},
  enlarge x limits=0.01,
  enlarge y limits=0.01,
]
\addplot table
[   x=X-Position,
  y=Probability
] 
{my.dat};
\end{axis}

Result 1

enter image description here

But questions remain:

do you have to do this on letter paper?
do the margins need to be this big?
wouldn't a logarhitmic plot be better, as the right side of the diagram is basically empty?

For comparison:

Code 2

\usepackage[margin=0.6in]{geometry}

...

\begin{semilogyaxis}
  ybar,
  xlabel=Character,
  ylabel=Probability,
  xtick=data,
  xticklabels from table={my.dat}{Character},
  bar width=5pt,
  width=0.95*\textwidth,
  x tick label style={font=\footnotesize,align=right,rotate=90},
  enlarge x limits=0.01,
  enlarge y limits=0.01,
]
\addplot table
[   x=X-Position,
  y=Probability
] 
{my.dat};
\end{semilogyaxis}

Result 2

enter image description here

As suggested by Jake, a scatter plot is probably better that a bar plot for this case:

Code 3

\begin{semilogyaxis}
[   scatter,
    scatter src=y,
    only marks,
  xlabel=Character,
  ylabel=Probability,
  xtick=data,
  xticklabels from table={my.dat}{Character},
  bar width=5pt,
  width=0.95*\textwidth,
  x tick label style={font=\footnotesize,align=right,rotate=90},
  enlarge x limits=0.01,
  enlarge y limits=0.02,
  grid=major,
  colormap={portal}{rgb255(0cm)=(255,128,0); rgb255(1cm)=(0,255,255)},
  colormap name=portal,
]
\addplot table
[   x=X-Position,
  y=Probability
] 
{my.dat};
\end{semilogyaxis}

Result 3

enter image description here

[Tex/LaTex] plotting two time series with bounds

This happens because PGFPlots only uses one "stack" per axis: You're stacking the second confidence interval on top of the first. The easiest way to fix this is probably to use the approach described in "Is there an easy way of using line thickness as error indicator in a plot?": After plotting the first confidence interval, stack the upper bound on top again, using stack dir=minus. That way, the stack will be reset to zero, and you can draw the second confidence interval in the same fashion as the first:

\documentclass{standalone}
\usepackage{pgfplots, tikz}

\usepackage{pgfplotstable}

\pgfplotstableread{
temps   y_h y_h__inf    y_h__sup    y_f y_f__inf    y_f__sup    

1   0.237340    0.135170    0.339511    0.237653    0.135482    0.339823    
2   0.561320    0.422007    0.700633    0.165871    0.026558    0.305184    
3   0.694760    0.534205    0.855314    0.074856    -0.085698   0.235411    
4   0.728306    0.560179    0.896432    0.003361    -0.164765   0.171487    
5   0.711710    0.544944    0.878477    -0.044582   -0.211349   0.122184    
6   0.671241    0.511191    0.831291    -0.073347   -0.233397   0.086703    
7   0.621177    0.471219    0.771135    -0.088418   -0.238376   0.061540    
8   0.569354    0.431826    0.706882    -0.094382   -0.231910   0.043146    
9   0.519973    0.396571    0.643376    -0.094619   -0.218022   0.028783    
10  0.475121    0.366990    0.583251    -0.091467   -0.199598   0.016664    
}{\table}

\begin{document}
\begin{tikzpicture}

    \begin{axis}
    % y_h confidence interval
    \addplot [stack plots=y, fill=none, draw=none, forget plot]   table [x=temps, y=y_h__inf]   {\table} \closedcycle;
    \addplot [stack plots=y, fill=gray!50, opacity=0.4, draw opacity=0, area legend]   table [x=temps, y expr=\thisrow{y_h__sup}-\thisrow{y_h__inf}]   {\table} \closedcycle;
    % subtract the upper bound so our stack is back at zero
    \addplot [stack plots=y, stack dir=minus, forget plot, draw=none] table [x=temps, y=y_h__sup] {\table};

    % y_f confidence interval
    \addplot [stack plots=y, fill=none, draw=none, forget plot]   table [x=temps, y=y_f__inf]   {\table} \closedcycle;
    \addplot [stack plots=y, fill=gray!50, opacity=0.4, draw opacity=0, area legend]   table [x=temps, y expr=\thisrow{y_f__sup}-\thisrow{y_f__inf}]   {\table} \closedcycle;

    % the line plots (y_h and y_f)    
    \addplot [stack plots=false, very thick,smooth,blue]  table [x=temps, y=y_h]   {\table};
    \addplot [stack plots=false, very thick,smooth,blue]  table [x=temps, y=y_f]   {\table};
    \end{axis}

\end{tikzpicture}
\end{document}

Best Answer

Related Solutions

[Tex/LaTex] Properly squish a pgfplot bar chart

Code 1

Result 1

Code 2

Result 2

Code 3

Result 3

[Tex/LaTex] plotting two time series with bounds

Related Question