Hands-On APIs for AI and Data Science

Chapter 11. Using APIs in Streamlit Data Apps

A simple app today is better than an over-designed app three months late.

Thiago Teixeira, cofounder of Streamlit

If you want to demonstrate your data science skills to other people, it’s hard to beat a data app—a web application displaying your data science models, graphs, charts, and spreadsheets. Sending a recruiter or potential client a link to a data app you created is a sure way to get their attention.

In this chapter, you will build a data app with Streamlit, an open source library that helps you create colorful and interactive web apps using Python. Streamlit handles all the complexity of the web interface so that you can focus on the data backend code, such as data files, pandas DataFrames, and APIs.

Streamlit apps can be deployed in a limited fashion for free on the Streamlit Community Cloud or paid web hosting platforms.

Engaging Users with Interactive Visualizations

Two main types of analytics products are tabular reports and visualizations. Tabular reports present rows and columns of data in a spreadsheet format, which provides a detailed view of data. Visualizations replace tabular data with charts, graphs, maps, and other images that give context and color to the data. Visualizations can communicate ideas that are hard to see in a spreadsheet.

Streamlit supports both tabular data and visualizations in a very interactive way. It provides built-in widgets, which are application controls such as sliders and select boxes ...