Skip to main content
Version: v1.0.0

Understand models

H2O MLOps lets you register individual experiments and group them as versions of a registered model to organize a collection of experiments efficiently.

Before deploying an experiment, you must register it in H2O MLOps. You can either register it as a new model or add it as a new version under an existing model. Choosing the second option creates a new version of the existing model.

A registered model is a collection of individual model versions. Registered models group related model versions that solve a specific problem. You can register new experiments and iterations as updated versions of the model.

A model version has a one-to-one relationship with an experiment within a given Workspace. When you're ready to serve your best experiment, register it as a model version.

note
  • When you register an experiment as a model, all data and metadata lineage are maintained.
  • Model versions can be served in multiple deployments. There is no limitation on the number of deployments a single model version can be a part of.
  • In any given workspace, an experiment can only be registered as one model version. This allows for a one-to-one mapping between an experiment and the model version.

Model schema​

Each model can be described by its input and output column names and their types. Knowing the model schema is essential for monitoring purposes. Currently, only models using the known schema can be deployed by MLOps.

Model schema is represented by the experiment metadata attached to the experiment. Deployer expects the model schema to be stored in the json_value of the input_schema and output_schema keys.

note

Natively supported Driverless AI MOJO2 and H2O-3 MOJO2 models are not required to contain the schema as the schema is an integral part of the MOJO2 artifacts.

Schema format​

The following example shows how model schema is formatted:

[{"name": "<column name>", "type": "<column type>"}, ... ]

Column types​

The following is a list of supported column types:

  • Boolean
  • Time64
  • Float32
  • Float64
  • Int32
  • Int64
  • String
note

Column type names are not case-sensitive.

Model type​

Model types enumerate types outputted by artifact processors. This indirection is included due to the fact that one particular artifact type can contain multiple internal artifacts, each of which may be consumed by different runtimes. One particular artifact type can be processed in different ways, producing different outputs consumable by different runtimes.

A model type defines what runtime can be used for artifact deployment.

To learn more about models in H2O MLOps, see:

Feedback