Questions

• Identify the best practices of an MLOps strategy

  • MLOps Workflow
  • Complete MLOps - End-to-End Pipeline

• Identify the advantages of using ML runtimes

  • Databricks Runtime for ML is optimized for machine learning workloads and comes pre-installed with many of the most popular ML libraries

• Identify how AutoML facilitates model/feature selection.

  • AutoML simplifies the process of applying machine learning to your datasets by automatically finding the best algorithm and hyperparameter configuration for you.
  • AutoML Fundamentals

• Identify the advantages AutoML brings to the model development process

  • AutoML Fundamentals

• Identify the benefits of creating feature store tables at the account level in Unity Catalog in Databricks vs at the workspace level

  • Feature Eng in Unity Catalog

• Create a feature store table in Unity Catalog

  Feature Eng in Unity Catalog

• Write data to a feature store table

• Train a model with features from a feature store table.

  Train Model from Feature Store in Unity Catalog

• Score a model using features from a feature store table.

  Scoring: batch inference

• Describe the differences between online and offline feature tables

  • The offline feature store is used for feature discovery, model training, and batch inference. It contains feature tables materialized as Delta tables.
  • An online store is a low-latency database used for real-time model inference. For a list of online stores that Databricks supports, see Third-party online stores.

• Identify the best run using the MLflow Client API.

  MLflow on Databricks

• Manually log metrics, artifacts, and models in an MLflow Run.

• Identify information available in the MLFlow UI

  • Catalog Explorer

• Register a model using the MLflow Client API in the Unity Catalog registry

• Identify benefits of registering models in the Unity Catalog registry over the workspace registry

  • Databricks recommends for governing and deploying models
  • Models in Unity Catalog extends the benefits of Unity Catalog to ML models, including centralized access control, auditing, lineage, and model discovery across workspaces. Models in Unity Catalog is compatible with the open-source MLflow Python client.

• Identify scenarios where promoting code is preferred over promoting models and vice versa

  MLOps

• Set or remove a tag for a model

  Model LifeCycle

• Promote a challenger model to a champion model using aliases

Summary