model

Ordering

Visibility

package dummy
package fs
FileSystem-based on Models.
FileSystem-based on Models.
It supports the following URIs.
- ../relative/path/to/model
- /absolute/path/to/model
- file:///absolute/path/to/model
- s3://bucket/path/to/model
- gs://bucket/path/to/model
- https://example.com/path/to/model
package mlflow
MLflow-based Models.
MLflow-based Models.
It supports the following URIs.
- mlflow://<runid>
- mlflow://<model> (latest version without a stage designation)
- mlflow://<model>/<stage> (latest version for given stage)
- mlflow://<model>/<version> (get the particular version of the registered model)
package testing
package tfhub
TFHub-based Models.
package torchhub
TorchHub-based Models.
TorchHub-based Models.
It supports the following URIs.
- torchhub:///<organization>/<project>/<model>

trait Catalog extends AnyRef
Catalog for SQL ML.
trait Model extends AnyRef
A Machine Learning Model in Rikai Catalog.
class ModelAlreadyExistException extends Exception
class ModelNameException extends Exception
class ModelNotFoundException extends Exception
class ModelRegistryAlreadyExistException extends Exception
class ModelResolveException extends Exception
case class ModelSpec(name: Option[String], uri: Option[String] = None, flavor: Option[String] = None, modelType: Option[String] = None, schema: Option[String] = None, options: Option[Map[String, String]] = None, preprocessor: Option[String] = None, postprocessor: Option[String] = None) extends Product with Serializable
Model Spec is used to pass the create model information to python ModelRegistry.
abstract class PyImplRegistry extends Registry with LazyLogging
trait Registry extends AnyRef
Model Registry Integrations.
class SimpleCatalog extends Catalog
A Simple Catalog in memory.
A Simple Catalog in memory. Used for local testing only.
class SparkUDFModel extends Model with SparkRunnable
A Model that can be turned into a Spark UDF.

Packages