-
Notifications
You must be signed in to change notification settings - Fork 29
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
knowpro: entity merging, refactoring (#760)
* Ongoing refactor * Removed currently unnecessary APIs * Entity and topic merging: * getDistinct* return ConcreteEntity[] and Topic[] * Reworked merging code; removed dependency on V1 code; temporary "compositeEntity" type * Future: query operator
- Loading branch information
Showing
10 changed files
with
225 additions
and
136 deletions.
There are no files selected for viewing
6 changes: 3 additions & 3 deletions
6
ts/packages/knowPro/src/conversation.ts → ts/examples/chat/src/memory/knowproCommon.ts
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,200 @@ | ||
| // Copyright (c) Microsoft Corporation. | ||
| // Licensed under the MIT License. | ||
|
|
||
| import { conversation as kpLib } from "knowledge-processor"; | ||
| import { collections, getTopK } from "typeagent"; | ||
| import { unionArrays } from "./collections.js"; | ||
| import { Scored, ScoredSemanticRef, SemanticRef, Topic } from "./dataFormat.js"; | ||
|
|
||
| export function facetValueToString(facet: kpLib.Facet): string { | ||
| const value = facet.value; | ||
| if (typeof value === "object") { | ||
| return `${value.amount} ${value.units}`; | ||
| } | ||
| return value.toString(); | ||
| } | ||
|
|
||
| export function mergeTopics( | ||
| semanticRefs: SemanticRef[], | ||
| semanticRefMatches: ScoredSemanticRef[], | ||
| topK?: number, | ||
| ): Scored<Topic>[] { | ||
| let mergedTopics = new Map<string, Scored<Topic>>(); | ||
| for (let semanticRefMatch of semanticRefMatches) { | ||
| const semanticRef = semanticRefs[semanticRefMatch.semanticRefIndex]; | ||
| if (semanticRef.knowledgeType !== "topic") { | ||
| continue; | ||
| } | ||
| const topic = semanticRef.knowledge as Topic; | ||
| const existing = mergedTopics.get(topic.text); | ||
| if (existing) { | ||
| if (existing.score < semanticRefMatch.score) { | ||
| existing.score = semanticRefMatch.score; | ||
| } | ||
| } else { | ||
| mergedTopics.set(topic.text, { | ||
| item: topic, | ||
| score: semanticRefMatch.score, | ||
| }); | ||
| } | ||
| } | ||
| if (topK !== undefined && topK > 0) { | ||
| return getTopK(mergedTopics.values(), topK); | ||
| } | ||
| return [...mergedTopics.values()]; | ||
| } | ||
|
|
||
| export function mergedEntities( | ||
| semanticRefs: SemanticRef[], | ||
| semanticRefMatches: ScoredSemanticRef[], | ||
| topK?: number, | ||
| ): Scored<kpLib.ConcreteEntity>[] { | ||
| return mergeScoredEntities( | ||
| getScoredEntities(semanticRefs, semanticRefMatches), | ||
| topK, | ||
| ); | ||
| } | ||
|
|
||
| type MergedEntity = { | ||
| name: string; | ||
| type: string[]; | ||
| facets?: MergedFacets | undefined; | ||
| }; | ||
|
|
||
| type MergedFacets = collections.MultiMap<string, string>; | ||
|
|
||
| function mergeScoredEntities( | ||
| scoredEntities: IterableIterator<Scored<kpLib.ConcreteEntity>>, | ||
| topK?: number, | ||
| ): Scored<kpLib.ConcreteEntity>[] { | ||
| let mergedEntities = new Map<string, Scored<MergedEntity>>(); | ||
| for (let scoredEntity of scoredEntities) { | ||
| const mergedEntity = concreteToMergedEntity(scoredEntity.item); | ||
| const existing = mergedEntities.get(mergedEntity.name); | ||
| if (existing) { | ||
| if (unionEntities(existing.item, mergedEntity)) { | ||
| if (existing.score < scoredEntity.score) { | ||
| existing.score = scoredEntity.score; | ||
| } | ||
| } | ||
| } else { | ||
| mergedEntities.set(mergedEntity.name, { | ||
| item: mergedEntity, | ||
| score: scoredEntity.score, | ||
| }); | ||
| } | ||
| } | ||
| let topKEntities = | ||
| topK !== undefined && topK > 0 | ||
| ? getTopK(mergedEntities.values(), topK) | ||
| : mergedEntities.values(); | ||
|
|
||
| const mergedConcrete: Scored<kpLib.ConcreteEntity>[] = []; | ||
| for (const scoredEntity of topKEntities) { | ||
| mergedConcrete.push({ | ||
| item: mergedToConcreteEntity(scoredEntity.item), | ||
| score: scoredEntity.score, | ||
| }); | ||
| } | ||
| return mergedConcrete; | ||
| } | ||
|
|
||
| /** | ||
| * In place union | ||
| */ | ||
| function unionEntities(to: MergedEntity, other: MergedEntity): boolean { | ||
| if (to.name !== other.name) { | ||
| return false; | ||
| } | ||
| to.type = unionArrays(to.type, other.type)!; | ||
| to.facets = unionFacets(to.facets, other.facets); | ||
| return true; | ||
| } | ||
|
|
||
| function concreteToMergedEntity(entity: kpLib.ConcreteEntity): MergedEntity { | ||
| return { | ||
| name: entity.name.toLowerCase(), | ||
| type: collections.lowerAndSort(entity.type)!, | ||
| facets: entity.facets ? facetsToMergedFacets(entity.facets) : undefined, | ||
| }; | ||
| } | ||
|
|
||
| function mergedToConcreteEntity( | ||
| mergedEntity: MergedEntity, | ||
| ): kpLib.ConcreteEntity { | ||
| const entity: kpLib.ConcreteEntity = { | ||
| name: mergedEntity.name, | ||
| type: mergedEntity.type, | ||
| }; | ||
| if (mergedEntity.facets && mergedEntity.facets.size > 0) { | ||
| entity.facets = mergedFacetsToFacets(mergedEntity.facets); | ||
| } | ||
| return entity; | ||
| } | ||
|
|
||
| function facetsToMergedFacets(facets: kpLib.Facet[]): MergedFacets { | ||
| const mergedFacets: MergedFacets = new collections.MultiMap< | ||
| string, | ||
| string | ||
| >(); | ||
| for (const facet of facets) { | ||
| const name = facet.name.toLowerCase(); | ||
| const value = facetValueToString(facet).toLowerCase(); | ||
| mergedFacets.addUnique(name, value); | ||
| } | ||
| return mergedFacets; | ||
| } | ||
|
|
||
| function mergedFacetsToFacets(mergedFacets: MergedFacets): kpLib.Facet[] { | ||
| const facets: kpLib.Facet[] = []; | ||
| for (const facetName of mergedFacets.keys()) { | ||
| const facetValues = mergedFacets.get(facetName); | ||
| if (facetValues && facetValues.length > 0) { | ||
| const facet: kpLib.Facet = { | ||
| name: facetName, | ||
| value: facetValues.join("; "), | ||
| }; | ||
| facets.push(facet); | ||
| } | ||
| } | ||
| return facets; | ||
| } | ||
|
|
||
| /** | ||
| * In place union | ||
| */ | ||
| function unionFacets( | ||
| to: MergedFacets | undefined, | ||
| other: MergedFacets | undefined, | ||
| ): MergedFacets | undefined { | ||
| if (to === undefined) { | ||
| return other; | ||
| } | ||
| if (other === undefined) { | ||
| return to; | ||
| } | ||
| for (const facetName of other.keys()) { | ||
| const facetValues = other.get(facetName); | ||
| if (facetValues) { | ||
| for (let i = 0; i < facetValues.length; ++i) { | ||
| to.addUnique(facetName, facetValues[i]); | ||
| } | ||
| } | ||
| } | ||
| return to; | ||
| } | ||
|
|
||
| function* getScoredEntities( | ||
| semanticRefs: SemanticRef[], | ||
| semanticRefMatches: ScoredSemanticRef[], | ||
| ): IterableIterator<Scored<kpLib.ConcreteEntity>> { | ||
| for (let semanticRefMatch of semanticRefMatches) { | ||
| const semanticRef = semanticRefs[semanticRefMatch.semanticRefIndex]; | ||
| if (semanticRef.knowledgeType === "entity") { | ||
| yield { | ||
| score: semanticRefMatch.score, | ||
| item: semanticRef.knowledge as kpLib.ConcreteEntity, | ||
| }; | ||
| } | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.