Introduction to H2O Driverless AI Python Client

Introduction to H2O Driverless AI Python Client

Course Level: Intermediate

H2O Driverless AI is a proprietary tool developed by to perform automatic machine learning without the need for coding. Automatic machine learning is the process of automating the tasks of applying machine learning to real-world problems. Driverless AI provides automatic feature engineering, model validation, model tuning, model selection and deployment and machine learning interpretability which can be used within a Python environment. By the end of this one-day course you will be able to use Driverless AI to create and analyse machine learning models for your data within Python.

Book: Introduction to H2O Driverless AI Python Client

Start Date:
Venue Details:

Course Details

  • Course Outline
  • Learning Outcomes
  • Materials
  • Prior Knowledge

Course Outline

  • Introduction to H2O Driverless AI: A brief introduction to H2O Driverless AI including a quick overview on automated machine learning
  • Getting started with Driverless AI Python Client: How to connect/create a Driverless AI instance
  • Loading and visualising data: How to create, load or view datasets; split into train/test/validation and visualise.
  • Launching an experiment: Discuss the different parameters that can be defined for an experiment for a regression or classification problem along with using these parameters when setting up and running an experiment.
  • Interpreting output: Explore feature engineering, summary information and model performance. Also download automated document report created using H2O AutoDoc.
  • Applying to new data: Apply a finalised model to new data (or test data) and assess fit.

Learning Outcomes

By the end of the course, participants will understand…

  • what is automated machine learning
  • how to create and connect to a Driverless AI instance
  • how to load a dataset into Driverless AI and set up an experiment
  • how to interpret model results and use the final model on new data.


Prior Knowledge

It will be assumed that participants are familiar with the Python programming language. For example, familiarity with dictionaries and use of functions and the libraries {numpy} and {pandas}. Some exposure to common statistical terms would be an advantage, but not essential. Attending the Introduction to Python and Programming with Python courses will provide a sufficient background for this course.

Attendee Feedback