Controlling view limits using margins and sticky_edges

The first figure in this example shows how to zoom in and out of a plot using margins instead of set_xlim and set_ylim. The second figure demonstrates the concept of edge "stickiness" introduced by certain methods and artists and how to effectively work around that.

import numpy as np
import matplotlib.pyplot as plt


def f(t):
    return np.exp(-t) * np.cos(2*np.pi*t)


t1 = np.arange(0.0, 3.0, 0.01)

ax1 = plt.subplot(212)
ax1.margins(0.05)           # Default margin is 0.05, value 0 means fit
ax1.plot(t1, f(t1))

ax2 = plt.subplot(221)
ax2.margins(2, 2)           # Values >0.0 zoom out
ax2.plot(t1, f(t1))
ax2.set_title('Zoomed out')

ax3 = plt.subplot(222)
ax3.margins(x=0, y=-0.25)   # Values in (-0.5, 0.0) zooms in to center
ax3.plot(t1, f(t1))
ax3.set_title('Zoomed in')

plt.show()
Zoomed out, Zoomed in

On the "stickiness" of certain plotting methods

Some plotting functions make the axis limits "sticky" or immune to the will of the margins methods. For instance, imshow and pcolor expect the user to want the limits to be tight around the pixels shown in the plot. If this behavior is not desired, you need to set use_sticky_edges to False. Consider the following example:

y, x = np.mgrid[:5, 1:6]
poly_coords = [
    (0.25, 2.75), (3.25, 2.75),
    (2.25, 0.75), (0.25, 0.75)
]
fig, (ax1, ax2) = plt.subplots(ncols=2)

# Here we set the stickiness of the axes object...
# ax1 we'll leave as the default, which uses sticky edges
# and we'll turn off stickiness for ax2
ax2.use_sticky_edges = False

for ax, status in zip((ax1, ax2), ('Is', 'Is Not')):
    cells = ax.pcolor(x, y, x+y, cmap='inferno', shading='auto')  # sticky
    ax.add_patch(
        plt.Polygon(poly_coords, color='forestgreen', alpha=0.5)
    )  # not sticky
    ax.margins(x=0.1, y=0.05)
    ax.set_aspect('equal')
    ax.set_title('{} Sticky'.format(status))

plt.show()
Is Sticky, Is Not Sticky