https://github.com/davidefiocco/streamlit-fastapi-model-serving

Simple web app example serving a PyTorch model using streamlit and FastAPI
https://github.com/davidefiocco/streamlit-fastapi-model-serving

docker-compose fastapi pycones pytorch streamlit

Last synced: about 1 month ago
JSON representation

Simple web app example serving a PyTorch model using streamlit and FastAPI

• Host: GitHub
• URL: https://github.com/davidefiocco/streamlit-fastapi-model-serving
• Owner: davidefiocco
• License: mit
• Created: 2020-06-20T18:57:39.000Z (over 4 years ago)
• Default Branch: main
• Last Pushed: 2024-02-12T12:31:46.000Z (9 months ago)
• Last Synced: 2024-09-29T17:40:55.471Z (about 1 month ago)
• Topics: docker-compose, fastapi, pycones, pytorch, streamlit
• Language: Python
• Homepage: https://davidefiocco.github.io/streamlit-fastapi-ml-serving
• Size: 46.9 KB
• Stars: 460
• Watchers: 7
• Forks: 89
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# streamlit-fastapi-model-serving

Simple example of usage of streamlit and FastAPI for ML model serving described on [this blogpost](https://davidefiocco.github.io/streamlit-fastapi-ml-serving) and [PyConES 2020 video](https://www.youtube.com/watch?v=IvHCxycjeR0).

When developing simple APIs that serve machine learning models, it can be useful to have _both_ a backend (with API documentation) for other applications to call and a frontend for users to experiment with the functionality.

In this example, we serve an [image semantic segmentation model](https://pytorch.org/hub/pytorch_vision_deeplabv3_resnet101/) using `FastAPI` for the backend service and `streamlit` for the frontend service. `docker compose` orchestrates the two services and allows communication between them.

To run the example in a machine running Docker and docker compose, run:

    docker compose build

    docker compose up

To visit the FastAPI documentation of the resulting service, visit http://localhost:8000/docs with a web browser.  

To visit the streamlit UI, visit http://localhost:8501.

Logs can be inspected via:

    docker compose logs

### Debugging

To modify and debug the app, [development in containers](https://davidefiocco.github.io/debugging-containers-with-vs-code) can be useful (and kind of fun!).