In this video we introduce MLflow & Model Serving on Databricks from a theoretical angle in a new playlist. With MLflow you can operationalize your ML workloads with experiment tracking & model registry via Unity Catalog. We explain how you can take a registered model and deploy on Databricks Model Serving as a REST endpoint.
NOTE: Next video in this playlist will be the hands-on if you want to skip there.
Video Resources
* Databricks Model Serving: https://www.databricks.com/product/model-serving
* Model Serving Intro Sample (Covered next video): https://github.com/RamVegiraju/databricks-samples/tree/master/ML/ModelServing
* MLFlow Intro: https://mlflow.org/docs/latest/ml/getting-started/
* Databricks Azure Account Setup: https://learn.microsoft.com/en-us/azure/databricks/getting-started/free-trial
Timestamps
0:00 Introduction
2:22 What is MLOps
4:30 Why MLFlow helps
7:12 Databricks Model Serving
12:26 UI Walkthrough
#databricks #mlflow #modelserving #machinelearning #mlops #azure
