Skip to content

Settings: Size Dependency

H2O Model Validation offers an array of settings for a Size Dependency test. Below, each setting is described in turn.

Training dataset

Defines the training dataset used in the Driverless AI (DAI) experiment that H2O Model Validation uses for the Size Dependency validation test.

Reference dataset

Defines the reference dataset used in the Driverless AI (DAI) experiment that H2O Model Validation uses for the Size Dependency validation test.

Note

By default, you cannot modify this setting; accordingly, a Size Dependency validation test can only run DAI experiments that use a test dataset.

Time column

Defines the time column of the train and reference dataset, which H2O Model Validation uses to perform time-based splits.

Number of splits

Defines the number of splits that H2O Model Validation performs on the training dataset to assess the dataset Size Dependency.

Note

The more splits, the better in general; it maximizes the test's explanatory power. Each split leads to an extra experiment run; this setting affects the runtime.

Remove validation experiments from DAI after finish

Determines if H2O Model Validation should delete the Driverless AI (DAI) artifacts, including experiments and datasets generated during the Size Dependency validation test. By default, H2O Model Validation checks this setting (enables it), and accordingly, H2O Model Validation deletes all DAI artifacts because they are no longer needed after the validation test is complete.