H2O-3 clients

These clients allow you to connect to and interact with H2O-3.

H2O Flow

H2O Flow is a Web based (GUI) user interface. It lets you interactively run your H2O-3 machine learning workflows and iteratively improve them. H2O Flow combines code execution, text, mathematics, plots, and rich media in a single document. See the documentation for H2O Flow.

R client

R users can use H2O-R library which internally uses H2O REST API calls to connect to H2O-3 (Server) and lets you run your H2O-3 workflow via R. See the documentation for the H2O-R Client.

Python client

Python users can connect to H2O-3 using the H2O Python package that internally uses H2O REST API calls to connect to H2O-3 (Server) and allows users to run their H2O-3 workflow via Python. See the documentation for the H2O-Python Client.

Sklearn support

Most H2O-3 estimators available in the H2O-Python client can also be used in the standard sklearn API. The h2o.sklearn module provides a collection of wrappers auto-generated on top of the original estimators and transformers, as well as on top of H2OAutoML.

See examples on how to integrate H2O-3 estimators into your Sklearn workflow.