Overview

In this tutorial, we’ll learn how to create interactive visualizations using Plotly Express, a high-level visualization library built on top of Plotly. Plotly creates web-based visualizations that allow users to hover, zoom, and interact with the data in ways that static visualizations cannot.

We’ll be using our now-familiar 1880 Utah Census data to explore various visualization techniques.

Setup

  • Use Github Desktop to sync up the shared class repository
  • The Utah Census dataset can be found in your week-09 folder utah-census-1880.csv
  • Create a new Jupyter Notebook file in the week-09 folder called plotly-intro.ipynb

First, let’s import the libraries we’ll need and load our dataset:

# Import libraries
import pandas as pd
import plotly.express as px

# Set pandas display options
pd.options.display.max_rows = 50

# Load the Utah 1880 Census data
utah_df = pd.read_csv('utah-census-1880.csv')

Let’s take a quick look at our data:

# Examine sample of rows
utah_df.sample(10)

Your First Plotly Visualization

Let’s start with something simple - a bar chart showing the number of people by county. We’ll first need to count the number of people in each county:

# Count the number of people in each county
county_counts = utah_df['county'].value_counts()

# Display the counts
county_counts

For Plotly (and many other visualization libraries), we need the data in a column format rather than as an index. We can use the reset_index() function to convert the index into a regular column:

# Apply reset_index() to convert the index (county names) into a regular column
county_counts = county_counts.reset_index()
county_counts.columns = ['county', 'count']

# Display the counts after reset_index
county_counts

The reset_index() function is essential when preparing grouped data for visualization, as it transforms the index values (in this case, county names) into a column that Plotly can use for the x-axis.

Now, let’s create our first Plotly bar chart:

# Create a simple bar chart
fig = px.bar(county_counts, x='county', y='count')

# Show the figure
fig.show()

Congratulations! You’ve created your first interactive Plotly visualization.

Customizing Your Visualizations

Now let’s customize our visualization to make it more informative and visually appealing.

Let’s improve our chart by adding a title and better axis labels:

# Add title and customize axis labels
fig = px.bar(
    county_counts,
    x='county',
    y='count',
    title='Population by County in Utah (1880)',  # Add a title
    labels={'county': 'County', 'count': 'Population'}  # Rename axis labels
)

# Display the chart
fig.show()

Now let’s make the chart more visually appealing by adding a different color scale (other options here) and using a different visual template for a cleaner white background.

# Add color and use a cleaner template
fig = px.bar(
    county_counts,
    x='county',
    y='count',
    title='Population by County in Utah (1880)',
    labels={'county': 'County', 'count': 'Population'},
    color='count',                     # Color bars by population
    color_continuous_scale='Inferno_r',  # Use a reversed color scale
    template='plotly_white'            # Use a clean white template
)

# Display the chart
fig.show()

Finally, we can adjust the layout to improve readability and appearance. We’re going to update the layout (.update_layout()) with several options to adjust the axis titles and the dimensions of the visualization.

# Add color and use a cleaner template
fig = px.bar(
    county_counts,
    x='county',
    y='count',
    title='Population by County in Utah (1880)',
    labels={'county': 'County', 'count': 'Population'},
    color='count',                     # Color bars by population
    color_continuous_scale='Inferno_r',  # Use a reversed color scale and the Inferno color scheme
    template='plotly_white'            # Use a clean white template
)

# Update layout with additional customizations
fig.update_layout(
    xaxis_title='County',                  # Customize x-axis title
    yaxis_title='Number of People',        # Customize y-axis title
    xaxis_tickangle=-45,                   # Rotate x-axis labels 45 degrees
    height=500,                            # Set chart height in pixels
    width=800,                             # Set chart width in pixels
    title_font=dict(size=22),              # Change title font size
    plot_bgcolor='white',                   # Set plot background color
    margin=dict(l=40, r=40, t=80, b=80),   # Adjust margins (left, right, top, bottom)
    showlegend=True,                       # Show the color scale legend
    legend_title_text='Population',        # Set legend title
    hoverlabel=dict(                       # Customize hover label appearance
        bgcolor="white",
        font_size=12,
        font_family="Arial")
)

# Display the chart
fig.show()

Create Your Own Visualization

Now it’s your turn to create a visualization using what you’ve learned. We’ll make a simple bar chart showing the distribution of the top 10 occupations in the Utah 1880 census:

# First, let's prepare our data
# Count the frequency of each occupation, take the top 10, and reset the index
top_occupations = utah_df['occupation'].value_counts().head(10).reset_index()
top_occupations.columns = ['occupation', 'count']

# Display the prepared data
top_occupations

Use what you’ve learned above to create an enhanced bar chart showing the top 10 occupations in the Utah Census that does the following across three code blocks:

  1. Create a bar chart using px.bar() with:

    • x-axis showing occupation names
    • y-axis showing the count
    • Bars colored by count (to highlight the most common occupations)

  2. Customize the chart with:

    • An appropriate title (e.g., “Top 10 Occupations in Utah (1880)”)
    • Clear axis labels
    • A different color scheme from the code above
    • A template OTHER than plotly_white

  3. Enhance the chart’s appearance with update_layout():

    • Rotate the x-axis labels if needed for readability
    • Adjust the chart dimensions (height: 400 and width: 800)
    • Customize the title font size to size 30
    • Set different margins from above
    • Add a legend and legend title
    • Customize the hover label appearance

Bonus

  • Export your chart as a static image and share it on the class Discord server

Resources for Learning More