Link axes with different Y-Scales

Alexander von Mach さんによって質問されました 2019 年 11 月 6 日

最新アクティビティ Adam Danz さんによって編集されました 2019 年 11 月 14 日

73 ビュー (過去 30 日間)

I'm trying to make a plot with 2 y-axes and wasn't able to use the yyaxis command since I want to change the uistack afterwards and it doesn't work with this function showhow. Now I have a new problem because of that. When I try to link the 2 axes, the y-scale of the left axis is changed to the values of the right one that are 10 times bigger.

Is there a way to link the axes to be able to "zoom" or change the parts visible without changing the axis values?

ax2 = axes('Position',[0.1300 0.1100 0.7750 0.8150]);
ax1 = axes('Position',[0.1300 0.1100 0.7750 0.8150]);
plot(ax1,[1 5 10],[1 5 10]); % Just random values not mine
plot(ax2,[1 5 10],[10 50 100]); % Just random values not mine
ax2.YAxisLocation = 'right';
set(ax1, 'Color', 'None');
ax2.XTick = [];
linkaxes([ax1 ax2]);

Thanks for your help

2 件のコメント

Adam Danz on 6 Nov 2019

"I'm trying to make a plot with 2 y-axes and wasn't able to use the yyaxis command since I want to change the uistack afterwards..."

There probably isn't a need to change the uistack. What exactly are you trying to do that involves the uistack? It would be cleaner to use the yyaxis function so if you could explain the problem you were having, maybe we could fix that instead of fixing the plan-b.

"Is there a way to link the axes to be able to "zoom" or change the parts visible without changing the axis values? "

It looks like the two datasets have different scales so you'll either need to

Scale your data so the range of y values match between the two datasets, or
don't link the axes but instead, carefully set the ylim() for each axis and the ytick.

To reiterate, I think the yyaxis solution is best and we could probably fix your uistack problem.

Alexander von Mach 2019 年 11 月 7 日

Thank you for your answer,

the problem I have with yyaxis right/left is, that the right axis is always at the top and I want the the plots of the left to be more visible since they're more important then the right one. I looked and it seems that this option isn't available in Matlab yet.

サインイン to comment.

1 件の回答

リンク

Answer by Adam Danz on 7 Nov 2019

Edited by Adam Danz on 14 Nov 2019

Accepted Answer

Currently (r2019b) the UI stack cannot be controlled with yyaxis axes which always results in right-axis objects at the top of the stack. If left-axis objects need to be on top of right-axis objects, use plotyy() which is the outdated version of yyaxis (as of r2016a). Just note that plotyy() may be removed in future releases (by then, hopefully uistack options will be available for yyaxis).

Here's a demo

clf()
sph = subplot(2,1,1);
[ax,h1,h2] = plotyy(sph,[0 10 20 30],[30 20 10 0],[0 10 20 30],[0 20 0 20]);
%                       [    x1    ],[    y1    ],[    x2    ],[    y2    ]
set([h1,h2],'LineWidth',3) % set line object properties
ax(1).Color = 'None';      % set axis 1 (left axis) to transparent
uistack(ax(1),'top')       % put axis 1 on top

5 件のコメント

表示 2 件の古いコメント

Adam Danz on 11 Nov 2019

If there are any further issues, let's discuss them here. It would be helpful if you provide a reproducible example.

Alexander von Mach on 14 Nov 2019

Hello Adam,

sorry for the late answer. I see that it works with your example but when I use simple plot-lines then it seems that the right axis is always on top.

figure()
ax(1) = subplot(2,1,1);
yyaxis(ax(1),'left')
plot([0 10 20 30],[0 20 0 20],'LineWidth',2)
yyaxis(ax(1),'right')
plot([0 10 20 30],[30 20 10 0],'LineWidth',2)
title('Right axis on top')
ax(2) = subplot(2,1,2);
yyaxis(ax(2),'right')
plot([0 10 20 30],[30 20 10 0],'LineWidth',2)
yyaxis(ax(2),'left')
plot([0 10 20 30],[0 20 0 20],'LineWidth',2)
title('Left axis on top')

Adam Danz 2019 年 11 月 14 日

@Alexander von Mach, there was a problem with my original answer, sorry to have led you astray. I've updated my answer with recommendations to use plotyy() which is an outdated function but it's a lot easier and cleaner than overlaying and linking 2 axes on your own.

サインイン to comment.

サインインしてこの質問に回答します。

現在この質問をフォロー中です

Alexander von Mach (view profile)