H2O License

[rfeigel]

How do I obtain an H2O license? Their website displays the license agreement, but no clear way to apply for a license.

[mlauber71]

Do you mean the open source version of H2O.ai or the commercial Driverless AI?

For the last one why not just contact their sales team? They offer a 21-day free trial. And as you might have noticed KNIME does support both the open source version and Driverless AI.

H2O.ai Documentation

h2oai/h2o-3/blob/master/LICENSE

[rfeigel]

Got a temporary license. Regression model runs to CSV Reader (Node #474) which tries to read from this location:

/Users/m_lauber/Dropbox/knime-workspace/hub/automl/kn_automl_h2o_regression_python/model/validate/H2O_AutoML_Regression_20200209_1841h_variable_importance.csv"/

Can you provide the file so I can install it locally? The classification model has a similar issue.

[mlauber71]

These points:

• you might have to reset the H2O workflow so it would use your own paths
• the workflow is based on a sample file of house prices and the validation is based on a few minutes of model time just to illustrate the results. It would do little to solve a real-world problem (that would take much more time and preparation of your data)
• the file is in the subfolder /model/validate of the workflow
• the workflow does use the open-source version of H2O.ai AutoML as described in the article below, there you will also find details about the validation which the mentioned file is a part (about variable importance)
• included in the workflow is a PDF in the /script/ subfolder with documentation “H2O.ai AutoML in KNIME for regression problems.pdf”
• if you want to use the Driverless AI you should follow their installation instructions and connect it to KNIME (check out the links provided or the examples 1|2)

H2O.ai AutoML in KNIME for regression problems

H2O.ai Automl - a powerful auto-machine-learning framework wrapped with KNIME It features various models like Random Forest or XGBoost along with Deep Learning. It has wrappers for R and Python but also could be used from KNIME. The results will be written to a folder and the models will be stored in MOJO format to be used in KNIME (as well as on a Big Data cluster via Sparkling Water). One major parameter to set is the running time the model has to test various models and do some hyperparam…

In general it would make sens to familiarize yourself with the concept of KNIME file handling and paths