This node is currently not available in KNIME v5.4 — instead we’re showing this page for KNIME v4.7. You can use the version menu in the title bar to permanently switch your preferred version. This will also show the link to the update site.

Run H2O Driverless AI Experiment

This node creates an experiment on the specified H2O Driverless AI server using a basic configuration. After the experiment finishes, the native GUI of Driverless AI for the finished experiment is available through the Interactive View.

Options

Connection

After the connection is provided to this node by the upstream connector, the following account parameters will not be displayed:

Username: Account username which the data will be uploaded to Driverless AI system.
Password: Account password.
H2O Driverless AI Location: Location of the Driverless AI server specifying the full URL including port number. Typical format is: http(s)://your-domain:12345 Note: the port number is 12345 by default but this can be reconfigured.
Connect & fetch data (button): After the connection to Driverless AI server is established, information about existing datasets and experiments on the remote server is fetched.

General Experiment Settings

Basic experiment settings that need to be specified. For more detail, see the documentation here.

Dataset: Dataset to train the model on. This can be pre-filled with a port connection to the Upload to Driverless AI node.
Target: The column to predict.
Prediction task: Whether to model the task as regression or classification. Classification is only available for target variables that are of a categorical (not continuous) type.
Name of the created experiment: The name of the new experiment. Note: the name must be unique on the server.
Accuracy: Accuracy controls the feature evolution and other aspects of model building process leading to attempts to build more or less accurate models.
Time: Time specifies the relative time for completing the experiment.
Interpretability: Interpretability specifies the relative interpretability for this experiment.

Input Ports

…: The connection that contains information about H2O Driverless AI authentication and the dataset that has been sent to H2O Driverless AI.

Output Ports

: Summary of the selected experiment.
: The H2O Driverless AI MOJO created by the experiment.

Popular Predecessors

Upload data to H2O Driverless AI100 %

Popular Successors

Views

Interactive View: H2O Driverless AI Experiment View: H2O Driverless AI Experiment View allowing refined work with the experiment such as interpreting the model.

Workflows

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.

Installation

To use this node in KNIME, install the extension KNIME H2O Driverless AI Integration from the below update site following our NodePit Product and Node Installation Guide:

v4.7

A zipped version of the software site can be downloaded here.

Plugin provider: KNIME AG

Plugin version: 1.0.0.v202112101518

On NodePit since: 2022-12-06

Last update: 2024-12-15

KNIME versions: v5.1, v4.7, v4.6, v4.5, v4.4, v4.3, v4.2

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!