Part 1 Conversation about Streamlit while walking

On a sunny afternoon, Rick and Chris were walking and chatting about Streamlit, a popular Python library for creating web applications. Their conversation flowed naturally, covering various aspects of this intriguing tool.

(This article originally appeared on 10/25/2024 on LinkedIN.)

Rick:“Hey Chris, I’ve been hearing a lot of buzz about Streamlit lately. So, what’s the scoop on it, especially when it comes to UI stuff? I’m toying with the idea of whipping up a CRUD app PoC with a slick interface. You think Streamlit’s the way to go for that kind of thing?”"Chris:“Oh, Streamlit is fantastic for that kind of project, Rick! It’s become quite popular for data apps and prototypes. The beauty of it is how it turns Python scripts into interactive web apps with minimal effort.”Rick:“That sounds promising. I am looking for a quick and dirty solution as a proof of concept. Hmmmm.. I wonder…. What kind of UI elements does it offer?”Chris:“Quite a range, actually. You’ve got your basic text elements like titles and headers, data display options for tables and metrics, and a variety of input widgets - buttons, sliders, text inputs, you name it. It even integrates well with data visualization libraries like Matplotlib and Plotly.”Rick:“Interesting. Are there any alternatives I should consider?”Chris:“Well, there’s Dash, which is also Python-based but more focused on analytical web applications. Or you could go the traditional route with Flask or Django paired with a frontend library, but that’s more complex.”Rick:“Got it. What about other languages? Any similar frameworks that simplify web development?”Chris:“Absolutely! If you’re into R, there’s Shiny. For Java developers, Vaadin is a great option. And if you’re looking to build desktop apps, Tauri and Electron are worth checking out.”Rick:“Thanks, that’s helpful. I think I will stick with Python for now. Can you walk me through the basics of how Streamlit works?”Chris:“Sure thing! Streamlit apps are basically Python scripts. You start by importing Streamlit, then use its functions to add widgets and layout elements. It’s reactive, so whenever an input changes, the script reruns from top to bottom, updating the app dynamically.”Rick:“That sounds straightforward. What about working with databases?”Chris:“While Streamlit doesn’t directly connect to databases, you can easily use Python’s database libraries. You’d typically use something like SQLAlchemy to connect to your database, run queries, and then display the results using Streamlit’s functions.”Rick:“And deploying a Streamlit app? How does that work?”Chris:“You’ve got several options there. Streamlit Cloud is the simplest - it connects directly to your GitHub repo. But you can also use services like Heroku, AWS Elastic Beanstalk, or even Docker if you prefer containerization.”Rick:“This has been really informative, Chris. Thanks a lot!”Chris:“Happy to help, Rick! If you have any more questions as you dive into Streamlit, don’t hesitate to ask. Happy coding!”

As their conversation came to a close, Rick felt equipped with a solid understanding of Streamlit and its capabilities, ready to embark on his new project with confidence.

An Introduction to Streamlit App Development

Streamlit is an open-source Python library that makes it easy to create and share beautiful, custom web apps for machine learning and data science. Whether you’re looking to build a quick prototype or a full-fledged data-driven application, Streamlit allows you to focus on the functionality and user interface without getting bogged down in traditional web development.

What is Streamlit?

Streamlit turns data scripts into shareable web apps in minutes. It’s as simple as writing Python scripts; Streamlit handles the rest. With an intuitive API, you can build interactive apps that can include widgets, charts, and data displays with minimal code.

Key Features of Streamlit

1. Simplicity and Speed

• Script-Based: Build apps by writing Python scripts.
• No Frontend Experience Needed: No need to know HTML, CSS, or JavaScript.
• Live Updates: Streamlit apps update in real-time as you edit and save your scripts.

2. Rich UI Components

Streamlit offers a variety of widgets and components to enhance your app:

• Text Elements: Titles, headers, markdown, and more for formatting text.
• Data Display: Tables, data frames (integrated with pandas), metrics, and code displays.
• Input Widgets: Buttons, sliders, text inputs, file uploaders, date pickers, and more.
• Media Elements: Display images, audio, and videos.
• Charts and Graphs: Integration with libraries like Matplotlib, Altair, Plotly, and more for data visualization.
• Layouts: Organize your app with columns, tabs, and expandable sections.

3. Reactive Programming Model

Streamlit reruns your entire script from top to bottom whenever a user interacts with a widget. This reactive model simplifies how you handle state and user interactions.

Basic Concepts and Structure

Getting Started

To start using Streamlit, install it via pip:

pip install streamlit

Writing Your First App

Create a Python script, say app.py, and import Streamlit:

import streamlit as st

st.title("My First Streamlit App")
st.write("Hello, world!")

Run your app with:

streamlit run app.py

Widgets and Interactivity

Streamlit provides various widgets to make your app interactive:

name = st.text_input("Enter your name")
st.write(f"Hello, {name}!")

age = st.slider("Select your age", 0, 100, 25)
st.write(f"You are {age} years old.")

Displaying Data

Streamlit works seamlessly with pandas DataFrames:

import pandas as pd

df = pd.DataFrame({
    'Name': ['Alice', 'Bob', 'Charlie'],
    'Age': [29, 34, 25]
})

st.dataframe(df)

Charts and Visualization

You can visualize data using built-in chart functions or integrate with popular libraries:

import numpy as np

chart_data = pd.DataFrame(
    np.random.randn(20, 3),
    columns=['a', 'b', 'c'])

st.line_chart(chart_data)

Working with Databases

While Streamlit doesn’t have built-in database support, you can use Python’s database libraries to interact with databases like PostgreSQL, MySQL, or MongoDB.

Connecting to a Database

Use libraries like sqlalchemy or psycopg2:

from sqlalchemy import create_engine

engine = create_engine('postgresql://username:password@localhost/dbname')
df = pd.read_sql('SELECT * FROM your_table', engine)
st.dataframe(df)

CRUD Operations

Create interactive widgets to perform CRUD operations:

if st.button("Add Entry"):
    # Code to insert into database
    pass

if st.button("Delete Entry"):
    # Code to delete from database
    pass

Authentication and Security

For apps requiring authentication, you can integrate Streamlit with identity providers like AWS Cognito.

Integrating AWS Cognito

While Streamlit doesn’t natively support authentication mechanisms, you can implement them using additional packages or custom code.

• Use OAuth2 or JWT Tokens: Secure your app by verifying tokens from Cognito.
• Middleware Solutions: Implement authentication checks at the start of your script.
• Third-party Libraries: Use libraries like streamlit-authenticator to simplify the process.

Deploying Your Streamlit App

Deployment is a crucial step to share your app with others.

1. Streamlit Community Cloud

• Quick and Easy: Ideal for prototypes and small apps.
• Free Tier Available: Deploy directly from a GitHub repo.
• Steps: Push your app code to GitHub. Sign up at Streamlit Community Cloud. Deploy your app by linking your GitHub repository.

2. Deploying on Azure

Deploying to Azure App Service is straightforward.

Prerequisites

• Azure account with an active subscription.
• Azure CLI installed locally.
• Your Streamlit app ready for deployment.

Deployment Steps

1. Log in to Azure
2. Create a Resource Group
3. Create an App Service Plan
4. Create a Web App
5. Prepare Your App
6. Configure the Startup Command
7. Deploy Your App
8. Access Your App

3. Other Deployment Options

• Heroku
• AWS Elastic Beanstalk
• Docker Containers
• Virtual Private Servers (VPS)

Each platform has its own set of instructions, but the general process involves setting up the environment, configuring dependencies, and running your Streamlit app.

Alternatives to Streamlit

If you’re looking for frameworks similar to Streamlit that abstract away web development complexities:

• Dash: Focuses on analytical applications; more structured.
• Gradio: Great for machine learning models; quick UI building.
• R Shiny: Ideal for R users to build interactive apps.
• Voila: Turns Jupyter Notebooks into standalone apps.
• Anvil: A platform for building full-stack web apps with Python.

Related links

• Github repo for streamlit tutorial
• Official Streamlit documentation
• Streamlit Gallery - for inspiration and examples
• Article: Streamlit Part 3 - Form Validation Part 1
• Streamlit Part 2: Layouts, Components and Graphs
• Part 1: Conversation about Streamlit while walking in the park
• Article: Streamlit Part 4 - Form Validation Part 2

Conclusion

Streamlit is a powerful tool that simplifies building interactive web applications in Python. With its intuitive API and rich set of features, you can focus on your data and logic rather than the intricacies of web development. Whether you’re a data scientist looking to share insights or a developer building a prototype, Streamlit offers a quick and effective solution.

Happy coding with Streamlit! If you have any questions or need more help, feel free to explore the official Streamlit documentation or contact the community.

About the Author

Rick Hightower is a seasoned software engineer and technology enthusiast with a passion for exploring cutting-edge tools in the world of data science and web development. With years of experience in the tech industry, Rick has a keen interest in simplifying complex processes and making technology more accessible to developers of all skill levels.

As an advocate for continuous learning, Rick enjoys sharing his knowledge and insights with the community. His exploration of Streamlit showcases his commitment to staying current with the latest trends in software development and his dedication to finding efficient solutions for building interactive data applications.

When not coding or writing about technology, Rick can be found enjoying nature walks and engaging in thought-provoking conversations about the future of tech with fellow enthusiasts.