Practical Machine Learning on Databricks

Packt via Coursera

Go to class Write review

Details

Go to class

Provider

Coursera
Pricing

Paid Course
Languages

English
Certificate

Certificate Available
Effort

10 weeks, 1 hour a week
Sessions

Self-Paced
Level

Advanced
Subtitles

English

Found in

Overview

Google, IBM & Meta Certificates – 40% Off

One plan covers every Professional Certificate on Coursera.

Unlock All Certificates

Machine learning teams today must build scalable, production-ready workflows that move seamlessly from experimentation to deployment and monitoring. This course provides a practical introduction to machine learning on Databricks, helping professionals leverage modern tools such as AutoML, MLflow, Feature Store, and serverless deployment to streamline ML operations and accelerate business outcomes. Through hands-on examples and real-world scenarios, you will learn how to create baseline models, manage feature engineering workflows, automate ML pipelines, and deploy models efficiently using Databricks. The course also explores model versioning, workflow orchestration, CI/CD automation, and model drift detection to help you maintain reliable and scalable machine learning systems in production environments. What sets this course apart is its strong focus on practical implementation using Databricks-native tools and workflows. By combining foundational ML concepts with enterprise-ready deployment and automation strategies, the course prepares you to tackle modern MLOps and machine learning engineering challenges with confidence. This course is ideal for data scientists, ML engineers, data engineers, and developers looking to transition to Databricks-based machine learning workflows. Learners should have prior experience with Python, machine learning concepts, and familiarity with Apache Spark fundamentals.

Syllabus

ML Process and Challenges

This module introduces the end-to-end machine learning lifecycle, highlighting common challenges faced in production environments and the importance of scalable, secure platforms. Learners will explore the roles involved in enterprise ML projects and discover how Databricks and Lakehouse architecture support collaboration and reproducibility. The module also examines how usability and complexity are balanced in modern ML platforms.

Overview of ML on Databricks

This module introduces learners to the foundational components of machine learning workflows on Databricks, including setting up a workspace, managing clusters, and utilizing key MLOps tools such as experiments and the feature store. Learners will gain practical knowledge on configuring environments and organizing ML development for scalable and collaborative projects.

Utilizing Feature Store

This module introduces learners to the Databricks Feature Store, guiding them through the process of registering feature tables and leveraging both offline and online stores for efficient feature management. Learners will gain hands-on experience with Delta tables and understand how to prepare features for model training and batch inference.

Understanding MLflow Components

This module introduces the core components of MLflow within the Databricks environment, focusing on experiment tracking, project management, and model registration. Learners will gain practical skills in standardizing and managing the machine learning lifecycle using MLflow tools. Hands-on examples will reinforce how to track and package ML models effectively.

Create a Baseline Model for Bank Customer Churn Prediction Using AutoML

This module guides learners through building a baseline machine learning model for predicting bank customer churn using Databricks AutoML. You will explore how to integrate MLflow and the Feature Store for streamlined model tracking and evaluation, and learn techniques for handling imbalanced datasets during model training.

Model Versioning and Webhooks

This module introduces the MLflow Model Registry, focusing on how to manage model versions and automate model lifecycle events using webhooks. Learners will discover how to streamline model deployment workflows and integrate external systems for real-time notifications and actions.

Model Deployment Approaches

This module guides learners through various strategies for deploying machine learning models on Databricks, including batch, streaming, and real-time inference. It also covers best practices for integrating custom Python libraries and managing dependencies to ensure scalable and efficient model delivery.

Automating ML Workflows Using the Databricks Jobs

This module introduces learners to the automation of machine learning workflows using Databricks Workflows and Jobs. You will discover how to schedule model retraining and testing, leverage Model Registry triggers, and integrate automation strategies for streamlined ML pipeline management.

Model Drift Detection for Our Churn Prediction Model and Retraining

This module introduces the concept of model drift in machine learning, focusing on how changes in data distributions can impact model performance. Learners will explore statistical methods, such as the chi-squared test, and practical tools in Databricks for detecting and addressing drift. By the end, you'll understand how to monitor, diagnose, and retrain models to maintain their effectiveness over time.

CI/CD to Automate Model Retraining and Re-Deployment.

This module introduces the principles and practices of automating machine learning model retraining and deployment using CI/CD pipelines within Databricks. Learners will explore MLOps workflows, deployment patterns, and the integration of MLflow for comprehensive model management. The content emphasizes real-world operational environments and strategies for effective model lifecycle automation.