Legend in bar plot

440 visualizaciones (últimos 30 días)
René Lampert
René Lampert el 3 de Jul. de 2021
Respondida: Greg LANGE el 19 de Oct. de 2022
Hello,
I wanna create a legend for a bar plot but I always get a warning message and it only shows one entitie of the legend. Here is a minimal code example
X=categorical({'small','medium','large'});
X=reordercats(X,{'small','medium','large'});
Y=[5 18 56];
TestL={'Mon','Tue','Wed'};
figure()
b=bar(X,Y);
b.FaceColor = 'flat';
b.CData(1,:)=[1 0 0];
b.CData(2,:)=[0 1 0];
b.CData(3,:)=[0 0 1];
legend(TestL)
And this is what i get from Matlab
What is wrong here ? I´m totally confused.
Thanks

Respuesta aceptada

dpb
dpb el 3 de Jul. de 2021
Editada: dpb el 3 de Jul. de 2021
A somewhat different approach to Walter's to generate the three needed bar handles -- use a 'stacked' plot with the elements on the diagonal, zero for the off-diagonal elements. 'stacked' creates a bar for each row whose heights are each the sums of the elements on the row. (And, need 'stacked' because otherwise passing an array to bar causes it to draw a 'grouped' bar.)
hB=bar(X,diag(Y,0),'stacked');
hLg=legend(TestL,'Location','northwest');
with default color produces
The bars can be customized as desired through the array of bar handles hB.
  3 comentarios
dpb
dpb el 3 de Jul. de 2021
set(hB,{'FaceColor'},{'r';'g';'b'})
hLg=legend(TestL,'Location','best')
will put where is least conflict with data -- if we do the above but swap the order to
hB=bar(flip(X),fliplr(diag(Y,0)),'stacked');
hLg=legend(TestL,'Location','best')
we get
where the legend has been moved automagically to the NE corner instead.
As an aside, I don't follow why this isn't the default legend behavior...
René Lampert
René Lampert el 4 de Jul. de 2021
thanks for your help

Iniciar sesión para comentar.

Más respuestas (2)

Walter Roberson
Walter Roberson el 3 de Jul. de 2021
legend() creates at most one entry for each graphics object. However, each call to bar() creates one graphics object, not one object for each group.
Create one extra bar() object for each item you want to legend(), and use nan as the data for that. legend() on those handles
X=categorical({'small','medium','large'});
X=reordercats(X,{'small','medium','large'});
Y=[5 18 56];
TestL={'Mon','Tue','Wed'};
figure()
b=bar(X,Y);
hold on
b.FaceColor = 'flat';
b.CData(1,:)=[1 0 0];
b.CData(2,:)=[0 1 0];
b.CData(3,:)=[0 0 1];
bh(1) = bar(nan,nan,'r');
bh(2) = bar(nan,nan,'g');
bh(3) = bar(nan,nan,'b');
legend(bh, string(X))
  2 comentarios
René Lampert
René Lampert el 3 de Jul. de 2021
thanks for your fast response...is it also possible to put the legend at another place, for instance at top left corner?
Walter Roberson
Walter Roberson el 3 de Jul. de 2021
X=categorical({'small','medium','large'});
X=reordercats(X,{'small','medium','large'});
Y=[5 18 56];
TestL={'Mon','Tue','Wed'};
figure()
b=bar(X,Y);
hold on
b.FaceColor = 'flat';
b.CData(1,:)=[1 0 0];
b.CData(2,:)=[0 1 0];
b.CData(3,:)=[0 0 1];
bh(1) = bar(nan,nan,'r');
bh(2) = bar(nan,nan,'g');
bh(3) = bar(nan,nan,'b');
legend(bh, string(X), 'location', 'northwest')

Iniciar sesión para comentar.


Greg LANGE
Greg LANGE el 19 de Oct. de 2022
What should I change to have my "Tuesday" to have the same color as my bar ? (purple)
What should I do to have my label as my picture with coordonates ?
Thank you in advance
figure (5)
%modifié
X=categorical({'small','medium','large'});
X=reordercats(X,{'small','medium','large'});
Y=[5 18 56];
TestL={'Mon','Tue','Wed'};
figure()
b=bar(X,Y);
hold on
b.FaceColor = 'flat';
b.CData(1,:)=[1 0 0];
%b.CData(2,:)=[0 1 0];
b.CData(2,:)=[0.4940 0.1840 0.5560];
b.CData(3,:)=[0 0 1];
bh(1) = bar(nan,nan,'r');
bh(2) = bar(nan,nan);%
bh(3) = bar(nan,nan,'b');
bh(2).CData=[0.4940 0.1840 0.5560];
legend(bh, TestL, 'location', 'best') %'northwest'
%legent at the top
xtips1 = b(1).XEndPoints;
ytips1 = b(1).YEndPoints;
labels1 = string(b(1).YData);
text(xtips1,ytips1,labels1,'HorizontalAlignment','center',...
'VerticalAlignment','bottom')
ylabel('Energie [kWh]')

Etiquetas

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by