matplotlib tutorial (15) nitta@tsuda.ac.jp¶

Chapter 15: Understand the relationship between the entire figure `Figure` and the coordinate system `Axes`¶

Basically it is recommended to generate as many Axes as you need at the same time as creating the figure.

However, if you want to create a weird Axes or have an Axes in a weird arrangement, first create a figure of the required size with figure().

matplotlib.figure.Figure(*args, **kwargs)
  [Parameters]
    figsize=(6.4, 4.8) :  (width, height) dimension in inches
    dpi=100 : dots per inch
  [Returns]
    Artist: the added artist

Then use figure.add_axes(rect=[left, bottom, width, height]) to place new Axes in any position you like.

matplotlib.figure.Figure.add_axes(self, *args, **kwargs)
  [Parameters]
    rect: [left, bottom, width, height]
    projection: projection types 
    sharex, sharey: share x or y axis
    label: label for the returned axes
  [Returns]
    axes

In the example below, it is specified as [left, bottom, width, height] = [0.1, 0.1, 0.8, 0.8] when adding Axes, so the blue Axes are placed like the figure below. In addition of the second Axes, it is specified as [left, bottom, width, height] = [0.6, 0.2, 0.4, 0.3], so the green Axes are place as follows.

In [ ]:

is_colab = 'google.colab' in str(get_ipython())   # for Google Colab

In [ ]:

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(-2, 3, 10)
y1 = x
y2 = x ** 2

fig = plt.figure(figsize=(5, 4))
ax = fig.add_axes([0.1, 0.1, 0.8, 0.8])
ax.scatter(x, y1)

ax2 = fig.add_axes([0.6, 0.2, 0.4, 0.3])
ax2.scatter(x, y2)

plt.show()

In [ ]: