Real Python Matplotlib Tutorial
Resources¶
import matplotlib.pyplot as plt
import numpy as np
np.random.seed(444)
-
One important big-picture matplotlib concept is its object hierarchy.
-
A “hierarchy” here means that there is a tree-like structure of matplotlib objects underlying each plot.
-
A Figure object is the outermost container for a matplotlib graphic, which can contain multiple Axes objects. One source of confusion is the name: an Axes actually translates into what we think of as an individual plot or graph (rather than the plural of “axis,” as we might expect).
fig, _ = plt.subplots()
type(one_tick := fig.axes[0].yaxis.get_major_ticks()[0])
-
Notice that we didn’t pass arguments to subplots() here. The default call is subplots(nrows=1, ncols=1)
fig, ax = plt.subplots()
type(ax)
-
We can call its instance methods to manipulate the plot similarly to how we call pyplots functions. Let’s illustrate with a stacked area graph of three time series
(rng := np.arange(50))
(rnd := np.random.randint(0, 10, size=(3, rng.size)))
(yrs := 1950 + rng)
fig, ax = plt.subplots(figsize=(5, 3))
ax.stackplot(yrs, rng + rnd, labels=["Eastasia", "Eurasia", "Oceania"])
ax.set_title("Combined debt growth over time")
ax.legend(loc="upper left")
ax.set_ylabel("Total debt")
ax.set_xlim(xmin=yrs[0], xmax=yrs[-1])
fig.tight_layout()
fig
Let’s look at an example with multiple subplots (Axes) within one Figure, plotting two correlated arrays that are drawn from the discrete uniform distribution:
(x := np.random.randint(low=1, high=11, size=50))
(y := x + np.random.randint(1, 5, size=x.size))
(data := np.column_stack((x, y)))
fig, (ax1, ax2) = plt.subplots(nrows=1, ncols=2, figsize=(8, 4))
ax1.scatter(x=x, y=y, marker="o", c="r", edgecolor="b")
ax1.set_title("Scatter: $x$ versus $y$")
ax1.set_xlabel("$x$")
ax1.set_ylabel("$y$")
ax2.hist(data, bins=np.arange(data.min(), data.max()), label=("x", "y"))
ax2.legend(loc=(0.65, 0.8))
ax2.set_title("Frequencies of $x$ and $y$")
ax2.yaxis.tick_right()
-
Text inside dollar signs utilizes TeXmarkup to put variables in italics.
-
Because we’re creating a “1x2” Figure, the returned result of plt.subplots(1, 2) is now a Figure object and a NumPy array of Axes objects. (You can inspect this with fig, axs = plt.subplots(1, 2) and taking a look at axs.)
fig
tuple(fig.axes[i] is ax for i, ax in zip(range(2), (ax1, ax2)))
-
Taking this one step further, we could alternatively create a figure that holds a 2x2 grid of Axes objects
fig, ax = plt.subplots(nrows=2, ncols=2, figsize=(7, 7))
-
Now, what is ax? It’s no longer a single Axes, but a two-dimensional NumPy array of them
type(ax)
ax
ax1, ax2, ax3, ax4 = ax.flatten()
To illustrate some more advanced subplot features, let’s pull some macroeconomic California housing data extracted from a compressed tar archive, using io, tarfile, and urllib from Python’s Standard Library.
import tarfile
from urllib.request import urlretrieve
from pathlib import Path
filepath, response = urlretrieve("http://www.dcc.fc.up.pt/~ltorgo/Regression/cal_housing.tgz")
filepath
with tarfile.open(name=filepath, mode='r') as archive:
housing = np.loadtxt(archive.extractfile("CaliforniaHousing/cal_housing.data"), delimiter=",")
housing
