[D] An Open Source Stack for Managing and Deploying ML Models – DVC & Cortex – Tutorial
In this tutorial, we’re going to use DVC to create a model capable of analyzing StackOverflow posts, and recognizing which ones are about Python. We are then going to deploy our model as a web API, ready to form the backend of a piece of production software.
DVC stores your model weights and training data in a centralized location, allowing collaborators to get started easily, while also tracking changes and ensuring an accurate version history.
As a final step in this tutorial, we’re going to integrate DVC with another open source tool—Cortex—that allows us to deploy DVC-generated models as web APIs, ready for production.