Skip to content

Private repository for the Vantage AI ML model productionization workshop

Notifications You must be signed in to change notification settings

BigDataRepublic/docker-workshop

Repository files navigation

Deploying your ML model

From notebook to production with FastAPI and Docker

This repository is for the Vantage AI ML model productionization workshop.

Instructions

To run the Docker container that hosts the FastAPI end-points, follow these instructions:

  • Clone this repository
  • Make sure the Docker daemon is running
  • Build the container image through copying the following command in your terminal: docker build -t docker_workshop/csgo_prediction_service .
  • Run the container with the following command: docker run -p 8000:8000 -t docker_workshop/csgo_prediction_service
  • Bonus: if you think those two commands are still to much to deploy your container, you can also use the Makefile and only run the make docker command

Et voilà! Your end-points should now be approachable via http://0.0.0.0:8000/.

Tip: visit http://0.0.0.0:8000/docs to make use of the Swagger UI for FastAPI and test your end-points!

Come join us!

Interested to work with us? Take a look at our vacancies! And make sure to follow us on LinkedIn!

About

Private repository for the Vantage AI ML model productionization workshop

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published