Databricks-ML-Associate cheat sheet

Exam at a glance

Domain weightings

How much of the exam each domain covers. Spend your study time in proportion — the heavier the domain, the more questions you'll see.

Key terms

Databricks Data Intelligence Platform

The Databricks Data Intelligence Platform is the unified lakehouse platform combining data engineering, analytics, and machine learning on Apache Spark and Delta Lake. The ML Associate exam's largest domain covers working within this platform for ML.

Databricks Runtime for ML

Databricks Runtime for Machine Learning is a cluster runtime that preinstalls common ML libraries such as scikit-learn, TensorFlow, PyTorch, XGBoost, and MLflow. The ML Associate exam expects you to choose it for ML workloads instead of the standard runtime.

MLflow

MLflow is the open-source platform, built into Databricks, for managing the ML lifecycle — tracking, projects, models, and the model registry. The ML Associate exam centers on MLflow tracking and the Unity Catalog model registry.

MLflow Tracking

MLflow Tracking is the component that logs parameters, metrics, artifacts, and models for each run so experiments are reproducible and comparable. The ML Associate exam covers autologging and comparing runs within an experiment.

Experiment

An Experiment is the MLflow container that groups related runs, each capturing the parameters and metrics of one training attempt. The ML Associate exam uses experiments to organize and compare model development.

Model Registry

The Model Registry is where MLflow models are versioned, aliased, and governed — on Databricks this is Models in Unity Catalog, which supersedes the legacy Workspace Model Registry. The ML Associate exam covers registering models and managing versions and aliases.

Unity Catalog

Unity Catalog is the Databricks governance layer for data and AI assets, providing centralized access control and lineage across workspaces. The ML Associate exam expects models and feature tables to be governed in Unity Catalog.

AutoML

AutoML is the Databricks feature that automatically trains and tunes candidate models for classification, regression, or forecasting and generates editable source notebooks. The ML Associate exam covers running AutoML and interpreting its outputs.

Feature Store

The Feature Store is the Databricks capability for creating, sharing, and serving curated feature tables (governed in Unity Catalog) for training and inference. The ML Associate exam covers feature engineering and retrieving features for models.

Spark ML

Spark ML (spark.ml) is the Apache Spark library for distributed machine learning using estimators, transformers, and pipelines on DataFrames. The ML Associate exam covers building Spark ML pipelines for models that must scale beyond a single node.

Pipeline

A Pipeline in Spark ML chains transformers and an estimator into a single reusable workflow that can be fit and applied consistently. The ML Associate exam uses pipelines to package feature transformation and model training together.

Hyperopt

Hyperopt is a hyperparameter-tuning library historically used on Databricks with SparkTrials to parallelize search across a cluster. It is now legacy — unmaintained after Databricks Runtime ML 16.4 LTS — with Optuna and Ray Tune the current recommended libraries.

pandas UDF

A pandas UDF (and applyInPandas) is a vectorized user-defined function that applies pandas logic across Spark partitions for scalable feature engineering or inference. The ML Associate exam uses it to distribute single-node model logic across a cluster.

Batch Inference

Batch Inference is scoring a large dataset at once by loading a registered MLflow model, often as a Spark UDF, and writing predictions back to a table. The ML Associate exam contrasts it with streaming and real-time serving.

Model Serving

Model Serving is the Databricks feature that exposes a registered model behind a REST endpoint for low-latency, real-time inference. The ML Associate exam covers real-time serving as one of the deployment paradigms alongside batch and streaming.