This content originally appeared on DEV Community and was authored by Eze Lanza
Join me with my guest William Arias, developer advocate at GitLab, as they show you step-by-step how to code and deploy a binary classifier using machine learning with open source tools. You’ll first prototype a solution in a Jupyter Notebook, then set up a basic CI pipeline to automate a machine learning app experimentation with MLFlow. Next, you’ll refactor the application (a bit) and create an API endpoint and simple UI for it to test manually (vibe testing). Before you’re done, you’ll learn how to combine everything and deploy the application to a runtime in the cloud using CI/CD principles.
In this session, you’ll learn how to:
• Prototype a binary classifier using machine learning in a Jupyter Notebook.
• Set up a basic continuous integration (CI) pipeline with MLFlow for app experimentation.
• Refactor the application to create an API endpoint and a simple user interface.
• Deploy the application to a cloud runtime using continuous deployment (CD) principles.
This content originally appeared on DEV Community and was authored by Eze Lanza
Eze Lanza | Sciencx (2024-08-12T23:05:59+00:00) S1E1: Code & Deploy: Build and Deploy an ML Binary Classifier. Retrieved from https://www.scien.cx/2024/08/12/s1e1-code-deploy-build-and-deploy-an-ml-binary-classifier/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.