Class Reference
IRIS for UNIX 2019.3
InterSystems: The power behind what matters   
Documentation  Search
  [ENSLIB] >  [%iKnow] >  [Classification] >  [Optimizer]
Private  Storage   

class %iKnow.Classification.Optimizer extends %RegisteredObject

This class automates selecting "appropriate" terms for a %iKnow.Classification.Builder. After pointing an Optimizer instance to the Builder object that needs optimization, use the LoadTermsArray and LoadTermsSQL methods to queue a large number of potentially interesting terms the Optimizer should test. Then invoke its Optimize method to let the Optimizer loop through the list of suggested terms automatically and add those terms having the highest positive impact on model accuracy (as measured according to ScoreMetric), removing terms that were already added to the model but turn out to have no significant positive impact on the model's accuracy.

See the individual property descriptions of their impact on the optimization process.

Inventory

Parameters Properties Methods Queries Indices ForeignKeys Triggers
19 17


Summary

Properties
AddCount AddWindowSize Builder CategoryWeights
CurrentClassifier CurrentScore CurrentTestId DomainId
MaximalScoreDecrease MetadataField MinimalScoreIncrease RemoveCount
RemoveStepRatio ScoreMetric TestSet Verbose

Methods
%AddToSaveSet %ClassIsLatestVersion %ClassName %ConstructClone
%DispatchClassMethod %DispatchGetModified %DispatchGetProperty %DispatchMethod
%DispatchSetModified %DispatchSetMultidimProperty %DispatchSetProperty %Extends
%GetParameter %IsA %IsModified %New
%NormalizeObject %ObjectModified %OriginalNamespace %PackageName
%RemoveFromSaveSet %SerializeObject %SetModified %ValidateObject
AddTerms Cleanup Initialize LoadTermsArray
LoadTermsSQL Optimize RemoveTerms SaveClassifier


Properties

• property AddCount as %Integer(MINVAL=1) [ InitialExpression = 1 ];
The number of terms to add during an AddTerms cycle. The top results according to RankScores will be added, as selected from the AddWindowSize terms tested in the cycle.
• property AddWindowSize as %Integer(MINVAL=0) [ InitialExpression = 0 ];
The number of terms to test in each round. If left at 0, this defaults to the number of cores the system has available, which should be most efficient.
• property Builder as %iKnow.Classification.Builder;
The builder object to be optimized.
• property CategoryWeights  [ MultiDimensional ];

If ScoreMetric is set to a 'Weighted*' value, the weights for each category are retrieved from this array, indexed by category name. If no category weight is set, it is assumed to be 0.

Note: Weights don't need to add up to 1.

• property CurrentClassifier as %String [ ReadOnly ];
The class name of the current "best" classifier. This value is set during Optimize, or as part of the AddTerms and RemoveTerms methods.
• property CurrentScore as %Double [ ReadOnly ];
The score of the current classifier. This value is updated by AddTerms and RemoveTerms.
• property CurrentTestId as %Integer [ ReadOnly ];
The key to %DeepSee.PMML.Utils.TempResult for the test results of CurrentClassifier.
• property DomainId as %Integer;
The domain using which the categorization model is being trained and tested. This assumes the value of the Builder's DomainId property when registering an IKnowBuilder instance as Builder, if not set explicitly.
• property MaximalScoreDecrease as %Double(MAXVAL=100,MINVAL=-100) [ InitialExpression = 0.05 ];
The maximal decrease in performance the optimizer should accept when trying to remove terms. If removing a term would imply a decrease larger than this figure, it will not be removed. A value of 1 means the maximal score decrease is 1%
• property MetadataField as %String;
The metadata field containing the actual category values to compare predictions against. This assumes the value of the Builder's MetadataField property when registering an IKnowBuilder instance as Builder, if not set explicitly.
• property MinimalScoreIncrease as %Double(MAXVAL=100,MINVAL=-100) [ InitialExpression = 0.1 ];
The minimal score increase % a term should ensure to be retained for further testing. If the score does not increase by at least this figure, it will be discarded from the list of terms to test. A value of 1 means the minimal score increase should be 1%
• property RemoveCount as %Integer(MINVAL=1) [ InitialExpression = 3 ];
The number of terms to remove in a "remove" cycle. Setting this value > 1 assumes the terms deemed irrelevant (and scheduled to be removed) don't influence one another much and removing more in a single cycle will not worsen performance much more than the individual performance changes of each term removal alone.
• property RemoveStepRatio as %Double(MAXVAL=1,MINVAL=0) [ InitialExpression = 0.1 ];

The ratio of RemoveTerms cycles vs AddTerms cycles. This should be a value between 0 and 1 (inclusive).

Note: Remove cycles take significantly longer than add cycles

• property ScoreMetric as %String(VALUELIST=",MacroFmeasure,MacroPrecision,MacroRecall,MicroFmeasure,MicroPrecision,MicroRecall,WeightedPrecision,WeightedRecall,WeightedFmeasure") [ InitialExpression = "MacroFmeasure" ];
The default accuracy metric to use for evaluating test results, as used by RankScores. If set to a 'Weighted*' value, the weights are retrieved from CategoryWeights.
• property TestSet as %iKnow.Filters.Filter;
The test set to validate model accuracy increases/decreases against.
• property Verbose as %String [ InitialExpression = 0 ];
If set to a boolean value, defines whether or not to write output to the current device during the Optimize method. If set to a string, it is treated as a global reference to which output needs to be written.

Methods

• method AddTerms(pCount As %Integer = -1, Output pAtEnd As %Boolean = 0) as %Status

This method does one round of processing, testing AddWindowSize candidate terms and selecting the best pCount terms according to RankScores, unless it wouldn't meet the MinimalScoreIncreas threshold.

If pCount < 0, it defaults to RemoveCount.

• method Cleanup() as %Status
This method clears the temporary artifacts the optimizer has created while optimizing, such as the CurrentClassifier class and CurrentTestId test results.
• method Initialize() as %Status
Initializes this Optimizer instance. This method is called automatically as part of Optimize
• method LoadTermsArray(ByRef pTerms, pListIndex As %Integer = 0) as %Status
Loads all terms from the supplied array. If pListIndex is non-zero, the term info is read from that index at each array position. If the term info itself is a list structure as well, it is interpreted as follows: pTerms(n) = $lb(term, type, negationpolicy, matchpolicy)
• method LoadTermsSQL(pSQL As %String) as %Status
Loads a list of candidate terms based on a SQL query. The query should return a column named "term" containing the term's value and may return columns named "type", "negation" and "match" to configure the type, negation and count policy for each term being retrieved, respectively.
• method Optimize(pMaxSteps As %Integer = 20) as %Status

In at most pMaxSteps steps, the current Builder will be optimized by testing, one at a time, the terms added through LoadTermsSQL and LoadTermsArray, judging which term works best for each test window by the results of RankScores (see also AddTerms). Every (1/) rounds, all terms in the dictionary so far will be tested for their contribution to the current model score and the lowest RemoveCount>RemoveStepRatio) rounds, all terms in the dictionary so far will be tested for their contribution to the current model score and the lowest RemoveCount terms will be removed (see also RemoveTerms).

At the end of the optimization process, in addition to Builder being updated, CurrentClassifier will contain the class name of the last test class used to achieve the best result and pTestId will point to the test results for that class.

• method RemoveTerms(pCount As %Integer = -1) as %Status

Test the impact of removing each term in the current model's TermDictionary individually. The pCount terms for which, after removing it, RankScores still returns the best score (which supposedly implies its contribution was minimial), will be removed from the TermDictionary, unless the decrease in performance surpasses MaximalScoreDecrease.

If pCount < 0, it defaults to RemoveCount.

• method SaveClassifier(pClassName As %String, pOverwrite As %Boolean = 0) as %Status
Saves the CurrentClassifier class to the desired pClassName, so it will not be removed after this Optimizer instance is dropped. If CurrentClassifier is not set or if the class no longer exists for other reasons, the current builder object will create a classifier class based on its current state.


Copyright (c) 2019 by InterSystems Corporation. Cambridge, Massachusetts, U.S.A. All rights reserved. Confidential property of InterSystems Corporation.