FacetGrid for multi-panel views in Data Analysis Python - Mini Project: Build & Apply

FacetGrid for multi-panel views

📖 Scenario: You work as a data analyst for a car company. You have data about different cars, including their type, fuel efficiency, and horsepower. You want to compare how horsepower relates to fuel efficiency for different types of cars.

🎯 Goal: Build a multi-panel plot using FacetGrid from the seaborn library to visualize the relationship between horsepower and fuel efficiency for each car type.

📋 What You'll Learn

Create a pandas DataFrame called cars with given data

Create a FacetGrid object with cars DataFrame, using car_type as the column facet

Map a scatter plot of horsepower vs mpg on the FacetGrid

Display the plot using plt.show()

💡 Why This Matters

🌍 Real World

Multi-panel plots help compare patterns across groups, such as car types, customer segments, or product categories.

💼 Career

Data analysts and scientists use FacetGrid to explore and present data insights clearly and efficiently.

1

Create the cars DataFrame

Create a pandas DataFrame called cars with these exact columns and values: car_type with ['Sedan', 'SUV', 'Sedan', 'SUV', 'Truck', 'Truck'], horsepower with [130, 250, 115, 300, 210, 230], and mpg with [30, 20, 35, 18, 15, 14].

Data Analysis Python

import pandas as pd
# Create the cars DataFrame with specified data
# Your code here

Hint

Use pd.DataFrame with a dictionary where keys are column names and values are lists of data.

2

Create the FacetGrid object

Import seaborn as sns and matplotlib.pyplot as plt. Then create a FacetGrid object called g using the cars DataFrame, setting col='car_type' to create one panel per car type.

Data Analysis Python

import pandas as pd
cars = pd.DataFrame({
    'car_type': ['Sedan', 'SUV', 'Sedan', 'SUV', 'Truck', 'Truck'],
    'horsepower': [130, 250, 115, 300, 210, 230],
    'mpg': [30, 20, 35, 18, 15, 14]
})

import seaborn as sns
import matplotlib.pyplot as plt
# Create FacetGrid object g with col='car_type'
# Your code here

Hint

Use sns.FacetGrid with cars and col='car_type'.

3

Map scatter plot on FacetGrid

Use the map method on the FacetGrid object g to plot a scatter plot with plt.scatter, mapping horsepower on the x-axis and mpg on the y-axis.

Data Analysis Python

import pandas as pd
cars = pd.DataFrame({
    'car_type': ['Sedan', 'SUV', 'Sedan', 'SUV', 'Truck', 'Truck'],
    'horsepower': [130, 250, 115, 300, 210, 230],
    'mpg': [30, 20, 35, 18, 15, 14]
})

import seaborn as sns
import matplotlib.pyplot as plt
g = sns.FacetGrid(cars, col='car_type')
# Map scatter plot of horsepower vs mpg on g
# Your code here

Hint

Use g.map(plt.scatter, 'horsepower', 'mpg') to plot points.

4

Display the plot

Use plt.show() to display the multi-panel scatter plot.

Data Analysis Python

import pandas as pd
cars = pd.DataFrame({
    'car_type': ['Sedan', 'SUV', 'Sedan', 'SUV', 'Truck', 'Truck'],
    'horsepower': [130, 250, 115, 300, 210, 230],
    'mpg': [30, 20, 35, 18, 15, 14]
})

import seaborn as sns
import matplotlib.pyplot as plt
g = sns.FacetGrid(cars, col='car_type')
g.map(plt.scatter, 'horsepower', 'mpg')
# Show the plot
# Your code here

Hint

Call plt.show() to display the figure window.

Practice

1. What is the main purpose of using FacetGrid in data visualization?

easy

A. To perform statistical tests on data

B. To create a single plot with multiple lines

C. To split data into multiple panels for easier comparison

D. To clean and preprocess data before plotting

Solution

1. Step 1: Understand FacetGrid functionality

   FacetGrid is designed to split data into multiple smaller plots based on categories.

2. Step 2: Identify the main use case

   This splitting helps compare subsets of data side by side visually.

3. Final Answer:

   To split data into multiple panels for easier comparison -> Option C

4. Quick Check:

   FacetGrid = multi-panel comparison [OK]

Hint: FacetGrid = split data into panels for comparison [OK]

Common Mistakes:

• Thinking FacetGrid creates a single combined plot
• Confusing FacetGrid with data cleaning tools
• Assuming FacetGrid performs statistical analysis

2. Which of the following is the correct way to create a FacetGrid with columns based on the 'species' column in a DataFrame df using seaborn?

easy

A. sns.FacetGrid(df, split='species')

B. sns.FacetGrid(df, row='species')

C. sns.FacetGrid(df, group='species')

D. sns.FacetGrid(df, col='species')

Solution

1. Step 1: Recall FacetGrid parameters

   The parameter to split data into columns is col.

2. Step 2: Match parameter with 'species'

   Using col='species' splits the grid into columns by species.

3. Final Answer:

   sns.FacetGrid(df, col='species') -> Option D

4. Quick Check:

   Use col= for columns in FacetGrid [OK]

Hint: Use col='column_name' to split by columns [OK]

Common Mistakes:

• Using 'group' or 'split' which are invalid parameters
• Confusing row and col parameters
• Missing quotes around column name

3. Given the following code, what will be the output type of g?

import seaborn as sns
import pandas as pd

df = pd.DataFrame({
    'category': ['A', 'A', 'B', 'B'],
    'value': [10, 15, 10, 20]
})
g = sns.FacetGrid(df, col='category')

medium

A. A FacetGrid object with two column panels for categories A and B

B. A DataFrame filtered by category

C. A matplotlib Figure object

D. A dictionary of plots keyed by category

Solution

1. Step 1: Understand FacetGrid return type

   Creating a FacetGrid returns a FacetGrid object that manages multiple plot panels.

2. Step 2: Check the col parameter effect

   Using col='category' splits the grid into columns for each unique category value, here 'A' and 'B'.

3. Final Answer:

   A FacetGrid object with two column panels for categories A and B -> Option A

4. Quick Check:

   FacetGrid object = multi-panel grid [OK]

Hint: FacetGrid returns a grid object, not data or dict [OK]

Common Mistakes:

• Expecting a DataFrame instead of a FacetGrid
• Thinking it returns a matplotlib Figure directly
• Assuming it returns a dictionary of plots

4. What is wrong with this code snippet?

import seaborn as sns
import pandas as pd

df = pd.DataFrame({
    'type': ['X', 'Y', 'X', 'Y'],
    'score': [5, 7, 6, 8]
})
g = sns.FacetGrid(df, col=type)
g.map(sns.histplot, 'score')

medium

A. DataFrame is missing required columns

B. The variable 'type' is not defined; it should be a string 'type'

C. sns.histplot cannot be used with FacetGrid

D. FacetGrid cannot use 'col' parameter

Solution

1. Step 1: Check parameter usage in FacetGrid

   The col parameter expects a string with the column name, not a variable.

2. Step 2: Identify the error in code

   Using col=type without quotes causes a NameError because 'type' is not defined as a variable.

3. Final Answer:

   The variable 'type' is not defined; it should be a string 'type' -> Option B

4. Quick Check:

   Use quotes for column names in FacetGrid [OK]

Hint: Always quote column names in FacetGrid parameters [OK]

Common Mistakes:

• Using variable names without quotes for column names
• Thinking FacetGrid can't use histplot
• Assuming DataFrame columns are missing

5. You want to create a multi-panel plot showing the distribution of 'total_bill' by 'day' and 'time' from the seaborn 'tips' dataset. Which code correctly creates a FacetGrid with rows for 'time' and columns for 'day' and plots histograms of 'total_bill'?

hard

A. g = sns.FacetGrid(tips, row='time', col='day')
g.map(sns.histplot, 'total_bill')

B. g = sns.FacetGrid(tips, col='time', row='day')
g.map(sns.histplot, 'total_bill')

C. g = sns.FacetGrid(tips, group='time', split='day')
g.map(sns.histplot, 'total_bill')

D. g = sns.FacetGrid(tips, row='total_bill', col='day')
g.map(sns.histplot, 'time')

Solution

1. Step 1: Identify correct row and column parameters

   Rows should be split by 'time' and columns by 'day' as per requirement.

2. Step 2: Confirm correct mapping of histogram

   Use g.map(sns.histplot, 'total_bill') to plot histograms of 'total_bill'.

3. Step 3: Check for invalid parameters

   Options with 'group' or swapped row/col are incorrect.

4. Final Answer:

   g = sns.FacetGrid(tips, row='time', col='day') g.map(sns.histplot, 'total_bill') -> Option A

5. Quick Check:

   row='time', col='day', map histplot on 'total_bill' [OK]

Hint: Use row= for rows, col= for columns, map plot on target column [OK]

Common Mistakes:

• Swapping row and col parameters
• Using invalid parameters like group or split
• Mapping plot on wrong column