bumplot
is a small Python package made to facilitate the creation of
bump charts using matplotlib and Bézier curves. It has high
customization capabilities too!
Bump charts are useful when the focus is on comparing relative rankings—who is ahead of whom—rather than the exact magnitude of the differences.
Check out the online documentation
import matplotlib.pyplot as plt
import pandas as pd
from bumplot import bumplot
data = pd.DataFrame(
{
"x": [2020, 2021, 2022, 2023],
"A": [10, 50, 20, 80],
"B": [40, 30, 60, 10],
"C": [90, 20, 70, 40],
}
)
fig, ax = plt.subplots(figsize=(8, 4))
bumplot(
x="x",
y_columns=["A", "B", "C"],
data=data,
curve_force=0.5,
plot_kwargs={"lw": 4},
scatter_kwargs={"s": 150, "ec": "black", "lw": 2},
colors=["#ffbe0b", "#ff006e", "#3a86ff"],
)
ax.legend()
ax.spines[["top", "right", "left", "bottom"]].set_visible(False)
ax.grid(alpha=0.4)
pip install bumplot
- Automatic ranking of your data (with support for both
pandas
andpolars
) - Accept categorical data for x-axis
- Infinite customization capabilities
- Easy control of the curvature intensity