Learn Combining Regression and Grids

The lmplot (linear model plot) is a figure-level function that combines regplot and FacetGrid.

While regplot is great for a single relationship, lmplot allows you to compare linear relationships across different categories. You can separate data by color (hue) or by splitting it into different subplots (col/row), making it powerful for answering questions like "Does the relationship between bill and tip change if the customer is a smoker?" .

Key Parameters

hue: separating the data by color and drawing a separate regression line for each group;
col / row: separating the data into distinct subplots;
markers: a list of symbols to distinguish groups visually (e.g., ['o', 'x']), which is helpful for accessibility.

Example

Here we compare the tips given during Lunch vs. Dinner. Notice how col splits the view, while hue compares smokers within each view.


              1234567891011121314151617
            
import seaborn as sns
import matplotlib.pyplot as plt

# Load dataset
df = sns.load_dataset('tips')

# Create a grid of regression plots
sns.lmplot(
    data=df,
    x='total_bill',
    y='tip',
    col='time',       # Split: Lunch vs Dinner
    hue='smoker',     # Color: Yes vs No
    palette='Set1'
)

plt.show()

Task

Swipe to start coding

Analyze the tips dataset to see how gender and smoking status affect tipping behavior.

Set the style to 'darkgrid'. Set the figure background color to 'lightpink'.
Create an lmplot using the tips dataset (df):
- Map 'total_bill' to x and 'tip' to y.
- Color the lines based on 'smoker' status (hue).
- Split the visualization into columns based on 'sex' (col).
- Use distinct markers: 'o' for the first group and 'x' for the second.
- Use the 'crest' palette.
Display the plot.

Solution

Everything was clear?

Thanks for your feedback!

Section 1. Chapter 22

single

Ask AI

Ask anything or try one of the suggested questions to begin our chat

Swipe to show menu