[FIXED] Beginner question, how to combine x and y ticks from 2 plots on a shared x and y axis

Issue

I try to get x and y ticks from 2 plots and share it on the x and y axis. With one plot it works

import matplotlib.pyplot as plt
import matplotlib.dates as md
from datetime import datetime

v1 = datetime.fromisoformat("2022-10-15T23:45:51.159618277Z")
v2 = datetime.fromisoformat("2022-10-16T23:49:51.159618277Z")
v3 = datetime.fromisoformat("2022-10-17T22:25:46.638699435Z")
t_time_values2 = [v1, v2, v3]
datenums = md.date2num(t_time_values2)
y = [30, 15, 45]

# plot lines
ax = plt.gca()
xfmt = md.DateFormatter('%Y-%m-%d %H:%M:%S.%f')
ax.xaxis.set_major_formatter(xfmt)
plt.plot(datenums, y, label="line 1")
ax.set_xticks(datenums)
ax.set_yticks(y)
plt.legend()
plt.show()

But now I try to combine 2 plots, and each x and y tick(from the plot) should be visible on the common x and y axis

I try it with this code, but unfortunately I can not merge the ticks from the 2 plots

import matplotlib.pyplot as plt
import matplotlib.dates as md
from datetime import datetime

v1 = datetime.fromisoformat("2022-10-15T23:45:51.159618277Z")
v2 = datetime.fromisoformat("2022-10-16T23:49:51.159618277Z")
v3 = datetime.fromisoformat("2022-10-17T22:25:46.638699435Z")
t_time_values1 = [v1, v2, v3]
datenums1 = md.date2num(t_time_values1)
y1 = [30, 15, 45]

#Plot2
v4 = datetime.fromisoformat("2022-10-15T05:45:51.159618277Z")
v5 = datetime.fromisoformat("2022-10-16T05:49:51.159618277Z")
v6 = datetime.fromisoformat("2022-10-17T02:25:46.638699435Z")
t_time_values2 = [v4, v5, v6]
datenums2 = md.date2num(t_time_values2)
y2 = [78, 25, 55]

# plot lines
ax = plt.gca()
xfmt = md.DateFormatter('%Y-%m-%d %H:%M:%S.%f')
ax.xaxis.set_major_formatter(xfmt)
plt.plot(datenums1, y1, label="line 1")
plt.plot(datenums2, y2, label="line 2")

ax.set_xticks(datenums1)
ax.set_xticks(datenums2)
ax.set_yticks(y1)
ax.set_yticks(y2)
plt.legend()
plt.show()

Solution

Please see how ax.set_xticks and ax.set_yticks work.

You are getting the current output because of your code: it is explicitly setting the x-ticks according to datenums2 because how you have asked it to at line number 28: ax.set_xticks(datenums2). This is undoing the expected result from the line written just before that, which is ax.set_xticks(datenums1).

Similar thing is happening with ax.set_yticks: it's only plotting yticks using y2.

There are different ways to handle this issue :

1. Let matplotlib take care of the tick labels and positions:

2. Define only y-ticks' locations and let maplotlib take care of y-ticks' labels, x-ticks' locations and their labels: 3. Define both x-ticks' and y-ticks' locations and let maplotlib take care of x-ticks' and y-ticks' labels. I am using different time but the dates are the same as provided in the question, please see the code below:

4. The best way to arrange xticks would be to properly format x-tick labels and let matplotlib take care of y-ticks:

Code for figure 1:

import matplotlib.pyplot as plt
import matplotlib.dates as md
from datetime import datetime

v1 = datetime.fromisoformat("2022-10-15T23:45:51")
v2 = datetime.fromisoformat("2022-10-16T23:49:51")
v3 = datetime.fromisoformat("2022-10-17T22:25:46")
t_time_values1 = [v1, v2, v3]
datenums1 = md.date2num(t_time_values1)
y1 = [30, 15, 45]

#Plot2
v4 = datetime.fromisoformat("2022-10-15T05:45:51")
v5 = datetime.fromisoformat("2022-10-16T05:49:51")
v6 = datetime.fromisoformat("2022-10-17T02:25:46")
t_time_values2 = [v4, v5, v6]
datenums2 = md.date2num(t_time_values2)
y2 = [78, 25, 55]

# plot lines
ax = plt.gca()
xfmt = md.DateFormatter('%Y-%m-%d %H:%M:%S.%f')
ax.xaxis.set_major_formatter(xfmt)
plt.plot(datenums1, y1, label="line 1")
plt.plot(datenums2, y2, label="line 2")

# Code above this comment will remain same for figure 2 and 3

plt.legend()
plt.show()

Code for figure 2:

# Add this in the code for figure 1
ax.set_yticks(y1 + y2)

y1 + y2 creates a list from the elements of y1 and y2.

Code for figure 3:

# Add these lines in the code for figure 1
ax.set_xticks(datenums1.tolist() + datenums2.tolist())
ax.set_yticks(y1 + y2)

datenums1 and datenums2 are numpy arrays, so if we try datenums1 + datenums2, it will just give another numpy array by doing an element-wise addition. That's why, we convert them to list first to get a list of x-ticks' locations.

Code for figure 4:

.
.
.
# plot lines
ax = plt.gca()

# xfmt = md.DateFormatter('%Y-%m-%d %H:%M:%S.%f')
# ax.xaxis.set_major_formatter(xfmt)

plt.plot(datenums1, y1, label="line 1")
plt.plot(datenums2, y2, label="line 2")

# Formatting of xticklabels 
xticks = datenums1.tolist() + datenums2.tolist()
labels = t_time_values1 + t_time_values2

def format_xlabels(x):
    fmt = '%Y-%m-%d %H:%M:%S.%f'
    label = x.strftime(fmt)
    label = label.rstrip("0") # Optional
    label = label.rstrip(".") # Optional
    return(label)

xlabels = [format_xlabels(i) for i in labels]
ax.set_xticks(xticks, labels=xlabels)

xlabels = ax.get_xticklabels()
plt.setp(xlabels, rotation=45, horizontalalignment='right')

plt.legend()
plt.tight_layout() # Important for xtick labels to show up in the figure
plt.show()

Please check the function my_format_function from here which I used to create the function format_xlabels in the code above.

Answered By - medium-dimensional