AutoML accelerates AI initiatives by providing methods and processes to make machine learning available for both experts and nonexperts alike, enhancing an organization’s ability to gain valuable insights from their data.

This technical white paper discusses automated machine learning, including the challenges of nonautomated AI model development and the benefits of automated machine learning. It provides an overview of the H2O Driverless AI product and presents a validated solution architecture for Driverless AI built on the Dell Validated Design for AI with VMware as the underlying infrastructure, including sizing recommendations for the infrastructure. It further describes how H2O Driverless AI for AutoML integrates with MLOps with cnvrg.io.

A practical real-world use case for sentiment analysis with NLP was developed and validated for this solution. The sentiment analysis is based on the analysis of tweets from the US Airline Sentiment dataset that assess positive, negative, and neutral comments from the public in real time. We also validated the solution using an image classification problem. These use cases are examples of how AutoML can be used to gain business value from a data stream.

AutoML with H2O Driverless AI on Dell infrastructure provides companies with an extensible and customizable data science platform that addresses the needs of various use cases for every enterprise in every industry.