ClassificationSolutionStore - Global

Enables storing and retrieving solutions.

The ClassificationSolutionStore API requires the Predictive Intelligence plugin (com.glide.platform_ml) and is provided within the sn_ml namespace.

ClassificationSolutionStore - add(Object mlSolution)

Adds a new solution object to the store and returns a unique name.

Note:

Label values do not need to be unique. For example, if you run this method with the same label 10 times, this method adds 10 different uniquely-named objects to the store.

Table 1. Parameters

Name	Type	Description
mlSolution	ClassificationSolution	ClassificationSolution() object to add to the store.

Table 2. Returns

Type	Description
String	System-generated solution name.

The following example shows how to add a solution to the store. Use ClassificationSolution - submitTrainingJob() to run the training job after adding it to the store.

// Create a dataset 
var myData = new sn_ml.DatasetDefinition({

  'tableName' : 'incident',
  'fieldNames' : ['assignment_group', 'short_description', 'description'],
  'encodedQuery' : 'activeANYTHING'

});

// Create a solution 
var mySolution = new sn_ml.ClassificationSolution({

  'label': "my solution definition",
  'dataset' : myData,
  'predictedFieldName' : 'assignment_group',
  'inputFieldNames':['short_description']

});

// Add the solution to the store to later be able to retrieve it.
var my_unique_name = sn_ml.ClassificationSolutionStore.add(mySolution);

ClassificationSolutionStore - deleteObject(String name)

Removes a specified solution object from the store.

Table 3. Parameters

Name	Type	Description
name	String	Name of the ClassificationSolution() object to be deleted.

Table 4. Returns

Type	Description
None

The following example shows how to delete a solution from the store.

sn_ml.ClassificationSolutionStore.deleteObject("ml_sn_global_global_solution");

ClassificationSolutionStore - get(String name)

Gets a solution object from the store.

Table 5. Parameters

Name	Type	Description
name	String	Name of a solution in the store.

Table 6. Returns

Type	Description
Object	ClassificationSolutionVersion object. Returns an error if the object does not exist.

The following example shows how to get a solution object from the store using the get() method and view its training status using the ClassificationSolution - getActiveVersion() and ClassificationSolutionVersion - getStatus() methods.

// Get status
var mlSolution = sn_ml.ClassificationSolutionStore.get('ml_incident_categorization');

gs.print(JSON.stringify(JSON.parse(mlSolution.getActiveVersion().getStatus(), null, 2)));

Output:

{
 "state":"solution_complete",
 "percentComplete":"100",
 "hasJobEnded":"true"
}

ClassificationSolutionStore - getAllNames(Object options)

Gets the names of all solution definition records in the store.

Table 7. Parameters

Name	Type	Description
options	Object	Options for restricting results within the specified properties. { "label": "String", "domainName": "String", "scope": "String" };
options.label	String	Optional. Label of your solution object.
options.domainName	String	Optional. Name of the domain for your solution object. Refer to Domain separation and Predictive Intelligence.
options.scope	String	Optional. Name of an application scope for your solution object.

Table 8. Returns

Type	Description
Array	List of strings representing solution object names in the store.

In the following example, the getAllNames() method returns a list of all names in the store.

gs.print(JSON.stringify(JSON.parse(sn_ml.ClassificationSolutionStore.getAllNames()), null, 2));

Output:

[
  "ml_x_snc_global_global_my_solution_definition_3",
  "ml_incident_assignment",
  "ml_x_snc_global_global_my_solution_definition",
  "ml_x_snc_global_global_my_solution_definition_2",
  "ml_sn_global_global_incident_service"
]

In the following example, the getAllNames() method returns only names associated with values set in the options parameter.

var options = {
  'label' : 'my solution definition',
  'domainName' : 'global',
  'scope' : 'global'
};
var solNames = sn_ml.ClassificationSolutionStore.getAllNames(options);
gs.print(JSON.stringify(JSON.parse(solNames), null, 2));

Output:

[
  "ml_x_snc_global_global_my_solution_definition"
]

ClassificationSolutionStore - update(String name, Object mlSolution)

Replaces an existing object in the store with the object passed as a parameter. The object name provided must be empty or match.

Table 9. Parameters

Name	Type	Description
name	String	Name of the solution to update.
mlSolution	ClassificationSolution	ClassificationSolution() object properties to update.

Table 10. Returns

Type	Description
None

The following example shows how to update a solution object in the store.

var solutionUpdate = new sn_ml.ClassificationSolution({
  'label': 'my solution definition',
  'dataset' : myData,
  'predictedFieldName' : 'assignment_group',
  'inputFieldNames': ['short_description']
});

sn_ml.ClassificationSolutionStore.update('ml_sn_global_global_incident_service', solutionUpdate);