Integrating Tensorflow Model to Streamlit


I have built a tensorflow object detection model that basically inputs videos and outputs the objects detected.

My biggest challenge has been in extracting and interpreting that data generated by the model.

Does Streamlit enable users to upload and deploy tensorflow models on its platform then perform analysis or does it just take the output (generated generated on my PC) and analyze it?

Hope that makes sense.

This looks like a great product!

Hi @Manhice! Thanks for the great question and welcome to the community! :slight_smile:

Currently, streamlit always runs locally. As an open source project, we collect usage statistics (which you can disable). We cannot see and do not store information contained in Streamlit apps.


  1. You can install Streamlit on a shared instance (like Amazon EC2, GCP, or Azure) in order to share a Streamlit app with you team. We encourage this! :muscle:

  2. Our upcoming Streamlit for Teams enables you to setup a server with one click, including on-prem or public hosting, access controls, and more. :rocket: If you’re interested, please sign up on that page.