matplotlib 3.5.1

>>> """
======
Spines
======

This demo compares:

- normal axes, with spines on all four sides;
- an axes with spines only on the left and bottom;
- an axes using custom bounds to limit the extent of the spine.
"""
... import numpy as np
... import matplotlib.pyplot as plt
... 
... 
... x = np.linspace(0, 2 * np.pi, 100)
... y = 2 * np.sin(x)
... 
... # Constrained layout makes sure the labels don't overlap the axes.
... fig, (ax0, ax1, ax2) = plt.subplots(nrows=3, constrained_layout=True)
... 
... ax0.plot(x, y)
... ax0.set_title('normal spines')
... 
... ax1.plot(x, y)
... ax1.set_title('bottom-left spines')
... 
... # Hide the right and top spines
... ax1.spines.right.set_visible(False)
... ax1.spines.top.set_visible(False)
... # Only show ticks on the left and bottom spines
... ax1.yaxis.set_ticks_position('left')
... ax1.xaxis.set_ticks_position('bottom')
... 
... ax2.plot(x, y)
... 
... # Only draw spine between the y-ticks
... ax2.spines.left.set_bounds(-1, 1)
... # Hide the right and top spines
... ax2.spines.right.set_visible(False)
... ax2.spines.top.set_visible(False)
... # Only show ticks on the left and bottom spines
... ax2.yaxis.set_ticks_position('left')
... ax2.xaxis.set_ticks_position('bottom')
... 
... plt.show()
...