Overview
Explore a comprehensive end-to-end machine learning workflow using MLflow on Databricks in this 40-minute tutorial. Learn how to leverage health data to predict life expectancy through a step-by-step process. Begin with data engineering in Apache Spark, followed by data exploration, model tuning, and logging using hyperopt and MLflow. Discover how the model registry governs model promotion and explore simple production deployment methods using MLflow as a job or dashboard. Gain insights into solving data science problems beyond model creation, including data cleaning, exploration, modeling, tuning, production deployment, and workflow management. Understand how Databricks' unified data analytics platform, powered by Apache Sparkâ„¢, accelerates innovation by bringing together data science, engineering, and business.
Syllabus
Introduction
Databricks
Data Engineering
Engineering
Koalas
Modeling
Model Registry
What happens next
Other things you can do
Moving to production
Interactive dashboard
Taught by
Databricks