How to Deploy Models Larger than 100MB on Streamlit

Author(s): Rifat Monzur

Originally published on Towards AI.

Three methods for deploying machine learning models larger than 100MB on Streamlit
Photo by Pat Whelen on Unsplash

For the last couple of months, I’ve been searching for an easy solution to create user interfaces and deploy my data science projects for the world to see (by ‘world,’ I mean myself and a couple of my buddies). Initially, I tried Flask, a Python micro-framework that’s very easy to learn. However, at the end of the day, you still need to use HTML, CSS, and JavaScript to design the user interface. I understand their necessity, but for my data science side projects, it’s a time-consuming process that I’d rather avoid. Additionally, I had to figure out the deployment process. I wanted something simpler.

This is where Streamlit shines. You don’t need to use anything other than Python (kinda!). You might write some markdown, but you can easily learn those on the go. If you really fancy, you could use some CSS to modify your UI, but that’s not really necessary. It is really up to you. For deployment, you just have to give access to your GitHub and indicate which repo you want to deploy. If your requirements.txt file has all your package names you used, you’re good to go.

Learning Streamlit was a breeze. I… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI