Skip to main content

%ML.Provider

Class %ML.Provider Extends %RegisteredObject [ Abstract, System = 4 ]

Base class for an ML provider

Parameters​

PROVIDERNAME​

Parameter PROVIDERNAME As %String;

Provider name

DEFAULTPROVIDER​

Parameter DEFAULTPROVIDER As %String = "AutoML";

Default provider name

SQLTYPE​

Parameter SQLTYPE As %String [ Constraint = "resultset,query", Flags = ENUM ] = "resultset";

Type of SQL TRAIN will operate on Options are "resutset" or "query" If the provider uses resultset, %Train is called with a resultset object for the query If the provider uses query, %TrainQyert is called with the SQL query text

Methods​

%CreateProvider​

ClassMethod %CreateProvider(name As %String = "", ByRef provider As %ML.Provider) As %Status

Factory method to create an instance of an ML provider by name

%GetProvider​

ClassMethod %GetProvider(name As %String = "", ByRef class As %String) As %Boolean

Look up a provider proper name

%GetName​

Method %GetName() As %String

Returns the provider name (the class name by default)

%GetTrainStatement​

Method %GetTrainStatement() As %SQL.Statement

Returns an instance of %SQL.Statement to be used for the training query

%GetDefaultSettings​

ClassMethod %GetDefaultSettings(ByRef settings As %DynamicObject)

Adds the default settings for the provider to the settings dynamic object

%OnClose​

Method %OnClose() As %Status

Make sure we clean up

%OnInit​

Method %OnInit() As %Status

Initialize a provider

%OnExit​

Method %OnExit() As %Status

Shutdown a provider

%BeginTraining​

Method %BeginTraining(model As %ML.Model, data As %SQL.StatementResult, trainingrun As %ML.TrainingRun, ByRef name As %String = "", ByRef trainkey) As %Status [ Abstract ]

Train an ML model

%WaitForTraining​

Method %WaitForTraining(ByRef trainkey, trainingrun As %ML.TrainingRun, ByRef trainedmodel As %ML.TrainedModel, timeoutMS As %Integer = -1) As %Status [ Abstract ]

Check for training complete