Charts with matplotlib

Numbers in a table are hard to feel. Ava's accountant wants to see the trend, and a picture will tell it in a second. Matplotlib is the most widely used plotting library in Python. A few lines turn your data into a clear chart, that you can save as an image, and put into a report or a website.

• Tables hide the trend
• A chart shows it in a second
• matplotlib: plots from code

A line chart

Import pyplot as plt. Call plot with the x values and the y values. Add a title, and labels for the axes, so the chart is understandable on its own. Then save the figure to a file, or show it on the screen. Here, Ava plots her weekly sales. The picture appears on the right.

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

weeks = [1, 2, 3, 4, 5, 6]
sales = [120, 150, 170, 160, 210, 260]
plt.plot(weeks, sales, marker="o", color="#3776ab")
plt.title("Weekly sales")
plt.xlabel("Week")
plt.ylabel("Dollars")
plt.grid(alpha=0.3)
plt.savefig("chart.png", dpi=110)

Matplotlib is building the font cache; this may take a moment.

Bar charts

Bar charts compare categories. Pass the names and the values to bar. Colours can be a list, one per bar, and you can add the value labels on top. Here, Ava compares the revenue of her best plants. The tallest bar is clear to see, even without reading a number.

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

plants = ["fern", "ivy", "rose", "cactus"]
revenue = [260, 180, 310, 140]
bars = plt.bar(plants, revenue, color=["#3776ab", "#6cb4f0", "#ffd43b", "#7be3a0"])
plt.bar_label(bars)
plt.title("Revenue by plant")
plt.savefig("chart.png", dpi=110)

Several charts in one figure

Use subplots to place several charts side by side. You get a figure, and an array of axes, and draw on each one. Here, a histogram shows how the order sizes are spread, and a scatter plot compares the price and the quantity. Tight layout prevents overlapping labels. Pandas has its own plot method, that calls matplotlib for you.

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt
import numpy as np

rng = np.random.default_rng(3)
sizes = rng.normal(30, 8, 200)
fig, (a, b) = plt.subplots(1, 2, figsize=(9, 3.6))
a.hist(sizes, bins=15, color="#3776ab")
a.set_title("Order sizes")
b.scatter(rng.random(40), rng.random(40), color="#ffd43b")
b.set_title("Price vs qty")
fig.tight_layout()
fig.savefig("chart.png", dpi=110)

Plot your own data

Your turn. Create a bar chart of three numbers, with a title. Save it to a file, and print a message when it is done. Ready? Here is one solution. A bar call, a title, a save, and a print.

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

plt.bar(["Mon", "Tue", "Wed"], [5, 9, 7])
plt.title("Sales this week")
plt.savefig("chart.png")
print("saved")

saved

Summary