Mastering Plotly in Python: Interactive Data Visualization Made Easy

Chapter 5: Plotly for Data Analysis and Visualization

🔹 1. Introduction

Once you’ve grasped the basics of Plotly and created interactive plots, the next step is using Plotly for more complex data analysis and visualization. In this chapter, we will focus on how to use Plotly for real-world data analysis, integrating it with Pandas DataFrames for exploratory data analysis (EDA) and data visualization.

This chapter will cover:

• Plotly with Pandas DataFrames
• Visualizing time-series data with Plotly
• Creating heatmaps for correlation matrices
• Generating box plots for distribution analysis
• Working with geospatial data using Plotly’s map visualizations

By the end of this chapter, you’ll be able to visualize complex datasets in a meaningful and interactive way, enabling you to uncover patterns and insights quickly.

🔹 2. Plotly with Pandas DataFrames

Pandas is the go-to library for data manipulation and analysis, and it integrates seamlessly with Plotly for visualization. In fact, Plotly Express works directly with Pandas DataFrames, allowing you to pass your data as a DataFrame and plot it with just a few lines of code.

✅ Plotting with Pandas DataFrames

Let’s start by creating a simple line chart from a Pandas DataFrame:

import plotly.express as px

import pandas as pd

# Sample data

data = {'Month': ['Jan', 'Feb', 'Mar', 'Apr', 'May'],

        'Sales': [100, 120, 150, 170, 200]}

df = pd.DataFrame(data)

# Create a line plot

fig = px.line(df, x='Month', y='Sales', title='Monthly Sales')

fig.show()

In this example, we pass a Pandas DataFrame df to px.line(), specifying the x and y columns. Plotly Express automatically handles the rest, creating an interactive line plot.

✅ Bar Plot from DataFrame

You can also create other types of plots, such as bar plots, directly from DataFrames:

fig = px.bar(df, x='Month', y='Sales', title='Sales per Month')

fig.show()

🔹 3. Visualizing Time-Series Data with Plotly

Time-series data is one of the most common types of data that you will encounter. Plotly makes it easy to visualize time-based trends with line charts and area charts.

✅ Plotting Time-Series Data

Here’s how you can plot time-series data using Plotly:

# Sample time-series data

data = {'Date': ['2021-01-01', '2021-02-01', '2021-03-01'],

        'Sales': [100, 120, 150]}

df = pd.DataFrame(data)

df['Date'] = pd.to_datetime(df['Date'])  # Convert 'Date' column to datetime

# Create a line plot

fig = px.line(df, x='Date', y='Sales', title='Sales Over Time')

fig.show()

In this example, the Date column is converted to datetime using Pandas to_datetime(), and Plotly is able to handle the time series visualization automatically.

✅ Time-Series with Multiple Series

You can also compare multiple time-series datasets on the same plot:

# Additional time-series data for comparison

data2 = {'Date': ['2021-01-01', '2021-02-01', '2021-03-01'],

         'Profit': [50, 60, 80]}

df2 = pd.DataFrame(data2)

df2['Date'] = pd.to_datetime(df2['Date'])  # Convert 'Date' column to datetime

# Create a plot with two time-series

fig = px.line(df, x='Date', y='Sales', title='Sales and Profit Over Time')

fig.add_scatter(x=df2['Date'], y=df2['Profit'], mode='lines', name='Profit')

fig.show()

This code creates a plot with Sales and Profit over time, displayed together for comparison.

🔹 4. Heatmaps for Correlation Matrices

A heatmap is an excellent way to visualize relationships between different variables in a dataset, particularly when examining correlations.

✅ Creating a Heatmap

Let’s visualize the correlation between several numerical variables using a heatmap:

import seaborn as sns

import plotly.figure_factory as ff

import pandas as pd

# Sample data

data = {'A': [1, 2, 3, 4, 5],

        'B': [5, 4, 3, 2, 1],

        'C': [1, 3, 5, 2, 4]}

df = pd.DataFrame(data)

# Calculate the correlation matrix

corr_matrix = df.corr()

# Create a heatmap

fig = ff.create_annotated_heatmap(z=corr_matrix.values, x=corr_matrix.columns.values, y=corr_matrix.index.values)

fig.show()

This creates a heatmap where the correlation values between columns are displayed as color gradients. The create_annotated_heatmap() function from Plotly’s figure_factory module is used to create the heatmap.

🔹 5. Box Plots for Distribution Analysis

Box plots are useful for visualizing the distribution of data, identifying outliers, and understanding the spread of the data.

✅ Creating a Box Plot

Box plots show the minimum, first quartile (Q1), median, third quartile (Q3), and maximum values of a dataset. Here’s how to create a box plot using Plotly:

import plotly.express as px

# Sample data

data = {'Category': ['A', 'A', 'B', 'B', 'C', 'C'],

        'Value': [10, 12, 15, 16, 25, 30]}

df = pd.DataFrame(data)

# Create a box plot

fig = px.box(df, x='Category', y='Value', title='Box Plot of Value by Category')

fig.show()

In this box plot, the Value column is displayed with categories on the x-axis. You can see the distribution and outliers for each category.

🔹 6. Working with Geospatial Data in Plotly

Plotly supports creating interactive maps for geospatial data visualization. These plots are ideal for visualizing geographic data, such as location-based data, sales by region, or heatmaps.

✅ Plotting Points on a Map

To visualize geographic data, you can use scatter geo plots:

import plotly.express as px

# Sample data for locations (latitude and longitude)

data = {'City': ['New York', 'Los Angeles', 'Chicago', 'Houston'],

        'Latitude': [40.7128, 34.0522, 41.8781, 29.7604],

        'Longitude': [-74.0060, -118.2437, -87.6298, -95.3698]}

df = pd.DataFrame(data)

# Create a scatter map plot

fig = px.scatter_geo(df, lat='Latitude', lon='Longitude', text='City', title='Cities in the US')

fig.show()

This will plot the cities on a map, showing their locations based on latitude and longitude.

🔹 7. Summary Table