Interchange location of y and z axis in a 3D plot

Question

By default, plotting a set of points (or whatever) in 3D with matplotlib, locates the z axis vertically, as seen here (code below):

I need to interchange the z and y axis, so that that y axis is shown vertically.

I've looked around but found no way to tell matplotlib to do this.

Add: I do not want to have to resort to a hack where I interchange the data and labels. This is a simple 3 points 3D plot, but I have to plot much more complicated surfaces. I'm looking for a general solution, not just something that works with scatter plots. A simple way to tell matplotlib to put the y axis vertically instead of the z axis is the clean way to do it.

MWE

import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D


fig = plt.figure()
ax = Axes3D(fig)

ax.scatter([0.2, 0.5, 0.8], [2.3, 0.47, 1.], [2.1, 5.3, 0.7])

ax.set_xlabel('x')
ax.set_ylabel('y')
ax.set_zlabel('z')
plt.show()

have you looked into the `zdir` argument to [`scatter`](http://matplotlib.org/mpl_toolkits/mplot3d/tutorial.html#mpl_toolkits.mplot3d.Axes3D.scatter)? — tmdavison, Jan 15 '16 at 14:17
I' m not familiar with the argument @tom. Keep in mind that I'm looking for a *general* solution, not just something that works on scatter plots. — Gabriel, Jan 15 '16 at 14:18
I have not used `zdir` before, would you like to post an answer? — Gabriel, Jan 15 '16 at 14:22
@Gabriel: Did you find a working solution? I'm having quite the same problem right now. — rtime, Nov 21 '16 at 19:54
Sadly, no @t.rathjen. The closest solution is the answer provided by innoSPG below. — Gabriel, Nov 21 '16 at 20:10
I'm also looking for a solution different from the hack. any updates no this? — martinako, Mar 08 '17 at 15:12

Bastian · Answer 1 · 2021-12-03T17:50:34.060

6

I don't think that this is currently possible. ax.view_init() would need to accept a third angle too. I opened an issue over at github, https://github.com/matplotlib/matplotlib/issues/14453#issue-452397120, let's hope someone is committed to implement this feature.

Update

The third view angle was recently implemented and has been merged into the matplotlib main branch: https://github.com/matplotlib/matplotlib/pull/21426

It should appear in Matplotlib 3.6.0.

edited Dec 03 '21 at 17:50

answered Jun 05 '19 at 09:24

Bastian

901
7
23

Thank you for opening the bug report Bastian! – Gabriel Jun 05 '19 at 12:42
4

This was a fun feature to implement, hope you all find it useful! – Scott Jan 05 '22 at 18:27

score 2 · Answer 2 · edited Jan 15 '16 at 14:11

2

One possibility is to interchange the position of y and z in the data and label accordingly. instead of

ax.scatter([0.2, 0.5, 0.8], [2.3, 0.47, 1.], [2.1, 5.3, 0.7])

use

ax.scatter([0.2, 0.5, 0.8], [2.1, 5.3, 0.7], [2.3, 0.47, 1.])

and label as ax.set_ylabel('z') and ax.set_zlabel('y')

edited Jan 15 '16 at 14:11

tmdavison

64,360
12
187
165

answered Jan 15 '16 at 14:11

innoSPG

4,588
1
29
42

3

Thanks inno, but this type of hack is precisely what I *don't* want to do. I'll add this to the question since it wasn't clear. – Gabriel Jan 15 '16 at 14:12

score 1 · Accepted Answer · answered Sep 16 '22 at 19:19

1

Matplotlib 3.6.0 is now out, and allows you to set this third view angle via the command ax.view_init(elev, azim, roll). See the official documentation.

answered Sep 16 '22 at 19:19

Scott

504
6
17

Interchange location of y and z axis in a 3D plot

3 Answers3

Linked