Skip to content

Deploying Your Streamlit App to Streamlit Cloud

Congratulations on setting up your project using the Streamlit Starter Kit! In this chapter, we'll take the next exciting step, deploying your bare-bones application to Streamlit Cloud. This will be the starting point of your application-building journey.

Streamlit Cloud is a platform that allows you to easily deploy, manage your Streamlit applications. By deploying your app to Streamlit Cloud, you'll be able to access it from anywhere, collaborate with others, and showcase your work to a wider audience.

In this chapter, we'll cover the following topics:

  • Creating a Streamlit Cloud account
  • Preparing your app for deployment
  • Connecting your GitHub repository to Streamlit Cloud
  • Configuring your app settings on Streamlit Cloud
  • Deploying your app and accessing it via a public URL
  • Making updates to your app and watch application refresh automatically in few seconds

By the end of this chapter, you'll have a live, publicly accessible Streamlit app that serves as a foundation for your application-building exercises. You'll be able to share the URL with others, gather feedback, and iterate on your app as you progress through the tutorial.

Let's dive in and get your app deployed to Streamlit Cloud!

If you are not on $TUTORIAL_HOME, naviaget to it and open the project in VS Code:

cd $TUTORIAL_HOME
code .

Install and Update Python packages

Note

Defining the right package versions will allow us to deploy the same application to SiS without any trouble.

Update the $TUTORIAL_HOME/requirements.txt to be like:

requirements.txt
snowflake-cli-labs>=3.1.0,<4.0.0
streamlit==1.35.0
pandas==2.0.3
numpy==1.24.3
scikit-learn==1.3.0

Creating a Python Virtual Environment

We will use miniconda to create a Python virtual environment and install all the packages.

Create an environment file $TUTORIAL_HOME/environment.yml to be like:

environment.yml
name: st_ml_app
channels:
  - snowflake
  - conda-forge
dependencies:
  - python=3.11
  - pip
  - pip:
    - -r requirements.txt

Run the following command to create the Python virtual environment:

conda env create -f environment.yml

And activate the environment:

conda activate st_ml_app

Pro Tip

Use direnv to de-clutter your local environment. You can create a Python virtual environment simply by creating a file named .envrc in your $TUTORIAL_HOME with the following content:

.envrc
layout_python

Then run direnv allow . to create a Python virtual environment in $TUTORIAL_HOME/.direnv/python-3.11

Application Update

Let us start with a small change to the application code.

Warning

For the entire tutorial, we will be making changes to the $TUTORIAL_HOME/streamlit_app.py file. The code listings will show the entire source with the changes highlighted. This approach helps avoid any copy/paste errors while doing the exercises.

Edit and update the $TUTORIAL_HOME/streamlit_app.py:

streamlit_app.py
1
2
3
4
5
import streamlit as st

st.title("🤖 Machine Learning App")

st.write("Welcome to world of Machine Learning with Streamlit.")
Commit the code: 

git commit -m "First deployment" streamlit_app.py

And push the code your remote(st-ml-app) Github repository.

Note

You can also do commit and push using VS Code. For the rest of the tutorial, commits and pushes are assumed implicit after each code update.

Deploying to Streamlit Community Cloud

To deploy the application navigate to https://streamlit.io, Sign-In and click Create app.

You will need the following details for deploying the app,

  • GitHub Repo name - <your-gh-user>/st-ml-app
  • Branch - main
  • Main file path - streamlit_app.py
  • App URL - choose a public url for your application easiest one to avoid is to use something like <your gh user>-ml-app.

Any commit and push to your repository will trigger a new application update. Give it a try!

🎉 Great! You have successfully deployed your bare-bones Streamlit app to Streamlit Cloud. Now you're ready to dive into the exciting world of building machine learning applications and discover how Streamlit can enhance that experience.

In the next chapter, we'll start transforming your starter app into a fully-fledged ML application.