1

Lets say I have: x1 = [1,2,3,4,5,6,7,8,9,10]; x2 = [1,3,10,60,500,10000,70000,500000,1000000,10000000]; y =[1,2,3,4,5,6,7,8,9,10] With matplotlib, how do I add x2 as a secondary axes (x1,y is the primary plot) when the only direct correlation between x1 and x2 is the index of the array? As in, there is no function to convert x1 to x2 (so I cannot use secondary_xaxis(pos, functions())? I've been trying secondary_xaxis for the past two days and just cannot get it to work. Also If i try to use set_xticks(x2) it bunches up the ticks but I want them directly corresponding to the x1 positions.

UPDATE:

my code looks like this:

self.x = [1,2,3,4,5,6,7,8,9,10]
self.y = [1,2,3,4,5,6,7,8,9,10]
x2 = [1,3,10,60,500,10000,70000,500000,1000000,10000000]
ax3 = ax.twiny()
ax.plot(self.x,self.y)
ax3.set_xticks(x2)
plt.show()

This gives the following graph. As you can see, the top ticks are not in line with where they should be (they should line up with the xticks on the bottom axis).

enter image description here

Aidan O'Farrell
  • 87
  • 6
  • Are you looking for something like [`twinx()`](https://matplotlib.org/stable/api/_as_gen/matplotlib.axes.Axes.twinx.html)? What is your code to produce the current plot? – t.o. Oct 28 '22 at 20:00
  • Yeah what i asked was not clear so i have updated my question with a little code and an image. @t.o. – Aidan O'Farrell Oct 28 '22 at 20:11

1 Answers1

1

From what I understand, you want the ticks of the top axis to align with the ticks at the bottom axis. For this, you need to define the amount of ticks on each axis and ensure they are the same. This is easily done by modifying the set_xticks() method for the ax object. Here is an example below

import numpy as np
import matplotlib.pyplot as plt

fig, ax = plt.subplots(ncols=1,nrows=1)
ax_o = ax.twiny()
N_ticks = 11


x1 = [1,2,3,4,5,6,7,8,9,10]
y = [1,2,3,4,5,6,7,8,9,10]
x2 = [1,3,10,60,500,10000,70000,500000,1000000,10000000]

ax.plot(x1,y) # Data for the bottom x-axis
ax.set_xticks(np.linspace(x1[0], x1[-1], N_ticks))
ax.set_xlim([x1[0], x1[-1]])

ax_o.plot(x2,y)# Data for the top x-axis
ax_o.set_xticks(x2) 
ax_o.set_xticks(np.linspace(x2[0], x2[-1], N_ticks))
ax_o.set_xlim([x2[0], x2[-1]])

plt.grid() # To show they are aligned
plt.show()

Which gives

result

Hope this helps

t.o.
  • 832
  • 6
  • 20
  • 1
    Thanks @t.o. this is exactly what I was looking for and actually stumbled acorss a very similar ticket with a similar answer (https://stackoverflow.com/questions/39540730/how-to-make-xticks-evenly-spaced-despite-their-value). Thanks for your help – Aidan O'Farrell Oct 28 '22 at 20:29
  • No worries! I forgot to plot the second data set, so the answer is updated once more. – t.o. Oct 28 '22 at 20:30