@wanshi-kg/wanshi 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/LICENSE +21 -0
- package/README.md +458 -0
- package/dist/__tests__/helpers.js +27 -0
- package/dist/__tests__/helpers.js.map +1 -0
- package/dist/cli/commands/export.command.js +99 -0
- package/dist/cli/commands/export.command.js.map +1 -0
- package/dist/cli/commands/index.js +22 -0
- package/dist/cli/commands/index.js.map +1 -0
- package/dist/cli/commands/inspectMerges.command.js +84 -0
- package/dist/cli/commands/inspectMerges.command.js.map +1 -0
- package/dist/cli/commands/metrics.command.js +196 -0
- package/dist/cli/commands/metrics.command.js.map +1 -0
- package/dist/cli/commands/process.command.js +82 -0
- package/dist/cli/commands/process.command.js.map +1 -0
- package/dist/cli/commands/watch.command.js +91 -0
- package/dist/cli/commands/watch.command.js.map +1 -0
- package/dist/cli/index.js +269 -0
- package/dist/cli/index.js.map +1 -0
- package/dist/cli/optionsToConfig.js +160 -0
- package/dist/cli/optionsToConfig.js.map +1 -0
- package/dist/config/index.js +59 -0
- package/dist/config/index.js.map +1 -0
- package/dist/config/legacyHints.js +113 -0
- package/dist/config/legacyHints.js.map +1 -0
- package/dist/config/schema.js +803 -0
- package/dist/config/schema.js.map +1 -0
- package/dist/config/ui.js +221 -0
- package/dist/config/ui.js.map +1 -0
- package/dist/core/DirectoryProcessor.js +725 -0
- package/dist/core/DirectoryProcessor.js.map +1 -0
- package/dist/core/adapters/IStructuredAdapter.js +3 -0
- package/dist/core/adapters/IStructuredAdapter.js.map +1 -0
- package/dist/core/adapters/SqliteAdapter.js +267 -0
- package/dist/core/adapters/SqliteAdapter.js.map +1 -0
- package/dist/core/adapters/StructuredAdapterRegistry.js +31 -0
- package/dist/core/adapters/StructuredAdapterRegistry.js.map +1 -0
- package/dist/core/adapters/index.js +20 -0
- package/dist/core/adapters/index.js.map +1 -0
- package/dist/core/checkpoint/CheckpointService.js +188 -0
- package/dist/core/checkpoint/CheckpointService.js.map +1 -0
- package/dist/core/checkpoint/index.js +18 -0
- package/dist/core/checkpoint/index.js.map +1 -0
- package/dist/core/corpus/CorpusAnalyzer.js +266 -0
- package/dist/core/corpus/CorpusAnalyzer.js.map +1 -0
- package/dist/core/corpus/CorpusProfileStore.js +92 -0
- package/dist/core/corpus/CorpusProfileStore.js.map +1 -0
- package/dist/core/corpus/index.js +21 -0
- package/dist/core/corpus/index.js.map +1 -0
- package/dist/core/corpus/normalizeGlossary.js +60 -0
- package/dist/core/corpus/normalizeGlossary.js.map +1 -0
- package/dist/core/corpus/relPath.js +52 -0
- package/dist/core/corpus/relPath.js.map +1 -0
- package/dist/core/corpus/termFrequency.js +86 -0
- package/dist/core/corpus/termFrequency.js.map +1 -0
- package/dist/core/cost/CostMeter.js +235 -0
- package/dist/core/cost/CostMeter.js.map +1 -0
- package/dist/core/cost/index.js +19 -0
- package/dist/core/cost/index.js.map +1 -0
- package/dist/core/cost/prices.js +38 -0
- package/dist/core/cost/prices.js.map +1 -0
- package/dist/core/cv/ObjectDetectionService.js +119 -0
- package/dist/core/cv/ObjectDetectionService.js.map +1 -0
- package/dist/core/di/ContainerFactory.js +670 -0
- package/dist/core/di/ContainerFactory.js.map +1 -0
- package/dist/core/di/DIContainer.js +103 -0
- package/dist/core/di/DIContainer.js.map +1 -0
- package/dist/core/di/index.js +19 -0
- package/dist/core/di/index.js.map +1 -0
- package/dist/core/errors/CustomErrors.js +342 -0
- package/dist/core/errors/CustomErrors.js.map +1 -0
- package/dist/core/errors/index.js +18 -0
- package/dist/core/errors/index.js.map +1 -0
- package/dist/core/export/KnowledgeGraphExportService.js +56 -0
- package/dist/core/export/KnowledgeGraphExportService.js.map +1 -0
- package/dist/core/export/index.js +19 -0
- package/dist/core/export/index.js.map +1 -0
- package/dist/core/export/strategies/GraphitiExportStrategy.js +115 -0
- package/dist/core/export/strategies/GraphitiExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/GraphvizDotExportStrategy.js +331 -0
- package/dist/core/export/strategies/GraphvizDotExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/IExportStrategy.js +3 -0
- package/dist/core/export/strategies/IExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/JsonExportStrategy.js +19 -0
- package/dist/core/export/strategies/JsonExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/JsonlExportStrategy.js +69 -0
- package/dist/core/export/strategies/JsonlExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/KblamExportStrategy.js +36 -0
- package/dist/core/export/strategies/KblamExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/LoraExportStrategy.js +46 -0
- package/dist/core/export/strategies/LoraExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/McpExportStrategy.js +67 -0
- package/dist/core/export/strategies/McpExportStrategy.js.map +1 -0
- package/dist/core/export/strategies/index.js +25 -0
- package/dist/core/export/strategies/index.js.map +1 -0
- package/dist/core/export/strategies/kbTriples.js +60 -0
- package/dist/core/export/strategies/kbTriples.js.map +1 -0
- package/dist/core/index.js +22 -0
- package/dist/core/index.js.map +1 -0
- package/dist/core/knowledge/KnowledgeGraphBuilder.js +627 -0
- package/dist/core/knowledge/KnowledgeGraphBuilder.js.map +1 -0
- package/dist/core/knowledge/MergeRecord.js +3 -0
- package/dist/core/knowledge/MergeRecord.js.map +1 -0
- package/dist/core/knowledge/canon/Canonicalizer.js +414 -0
- package/dist/core/knowledge/canon/Canonicalizer.js.map +1 -0
- package/dist/core/knowledge/canon/index.js +18 -0
- package/dist/core/knowledge/canon/index.js.map +1 -0
- package/dist/core/knowledge/contradiction/HeuristicContradictionChecker.js +92 -0
- package/dist/core/knowledge/contradiction/HeuristicContradictionChecker.js.map +1 -0
- package/dist/core/knowledge/contradiction/LlmContradictionChecker.js +52 -0
- package/dist/core/knowledge/contradiction/LlmContradictionChecker.js.map +1 -0
- package/dist/core/knowledge/contradiction/index.js +19 -0
- package/dist/core/knowledge/contradiction/index.js.map +1 -0
- package/dist/core/knowledge/grounding/KeywordGroundingChecker.js +33 -0
- package/dist/core/knowledge/grounding/KeywordGroundingChecker.js.map +1 -0
- package/dist/core/knowledge/grounding/MiniCheckGroundingChecker.js +82 -0
- package/dist/core/knowledge/grounding/MiniCheckGroundingChecker.js.map +1 -0
- package/dist/core/knowledge/grounding/index.js +20 -0
- package/dist/core/knowledge/grounding/index.js.map +1 -0
- package/dist/core/knowledge/grounding/verbalize.js +38 -0
- package/dist/core/knowledge/grounding/verbalize.js.map +1 -0
- package/dist/core/knowledge/images/imageMetaGraph.js +136 -0
- package/dist/core/knowledge/images/imageMetaGraph.js.map +1 -0
- package/dist/core/knowledge/index.js +20 -0
- package/dist/core/knowledge/index.js.map +1 -0
- package/dist/core/knowledge/merging/KnowledgeMerger.js +624 -0
- package/dist/core/knowledge/merging/KnowledgeMerger.js.map +1 -0
- package/dist/core/knowledge/references/ReferenceResolver.js +184 -0
- package/dist/core/knowledge/references/ReferenceResolver.js.map +1 -0
- package/dist/core/knowledge/references/citations/CitationEvidenceProcessor.js +401 -0
- package/dist/core/knowledge/references/citations/CitationEvidenceProcessor.js.map +1 -0
- package/dist/core/knowledge/references/citations/CitationResolver.js +95 -0
- package/dist/core/knowledge/references/citations/CitationResolver.js.map +1 -0
- package/dist/core/knowledge/references/citations/GrobidClient.js +143 -0
- package/dist/core/knowledge/references/citations/GrobidClient.js.map +1 -0
- package/dist/core/knowledge/references/citations/TitleIdResolver.js +101 -0
- package/dist/core/knowledge/references/citations/TitleIdResolver.js.map +1 -0
- package/dist/core/knowledge/references/web/FetchCacheService.js +114 -0
- package/dist/core/knowledge/references/web/FetchCacheService.js.map +1 -0
- package/dist/core/knowledge/references/web/GatedFetcher.js +228 -0
- package/dist/core/knowledge/references/web/GatedFetcher.js.map +1 -0
- package/dist/core/knowledge/references/web/WebReferenceProcessor.js +164 -0
- package/dist/core/knowledge/references/web/WebReferenceProcessor.js.map +1 -0
- package/dist/core/knowledge/search/KnowledgeGraphSearch.js +261 -0
- package/dist/core/knowledge/search/KnowledgeGraphSearch.js.map +1 -0
- package/dist/core/knowledge/vocabulary.js +162 -0
- package/dist/core/knowledge/vocabulary.js.map +1 -0
- package/dist/core/llm/EmbeddingService.js +113 -0
- package/dist/core/llm/EmbeddingService.js.map +1 -0
- package/dist/core/llm/OllamaService.js +146 -0
- package/dist/core/llm/OllamaService.js.map +1 -0
- package/dist/core/llm/OpenAICompatibleService.js +190 -0
- package/dist/core/llm/OpenAICompatibleService.js.map +1 -0
- package/dist/core/llm/OpenAIEmbeddingService.js +129 -0
- package/dist/core/llm/OpenAIEmbeddingService.js.map +1 -0
- package/dist/core/llm/embeddingUtils.js +25 -0
- package/dist/core/llm/embeddingUtils.js.map +1 -0
- package/dist/core/llm/index.js +23 -0
- package/dist/core/llm/index.js.map +1 -0
- package/dist/core/llm/prompts/PromptManager.js +388 -0
- package/dist/core/llm/prompts/PromptManager.js.map +1 -0
- package/dist/core/llm/prompts/PromptTemplateEngine.js +257 -0
- package/dist/core/llm/prompts/PromptTemplateEngine.js.map +1 -0
- package/dist/core/llm/prompts/templates/partials/examples/EXAMPLE_STYLE_GUIDE.md +84 -0
- package/dist/core/llm/prompts/templates/partials/examples/article.md +187 -0
- package/dist/core/llm/prompts/templates/partials/examples/code.md +229 -0
- package/dist/core/llm/prompts/templates/partials/examples/communication.md +205 -0
- package/dist/core/llm/prompts/templates/partials/examples/documentation.md +262 -0
- package/dist/core/llm/prompts/templates/partials/examples/financial.md +157 -0
- package/dist/core/llm/prompts/templates/partials/examples/legal.md +153 -0
- package/dist/core/llm/prompts/templates/partials/examples/logs.md +127 -0
- package/dist/core/llm/prompts/templates/partials/examples/medical.md +218 -0
- package/dist/core/llm/prompts/templates/partials/examples/notes.md +201 -0
- package/dist/core/llm/prompts/templates/partials/examples/research.md +208 -0
- package/dist/core/llm/prompts/templates/partials/examples/tabular.md +178 -0
- package/dist/core/llm/prompts/templates/partials/examples/transcript.md +204 -0
- package/dist/core/llm/prompts/templates/partials/retrieved-context.hbs +18 -0
- package/dist/core/llm/prompts/templates/v1/system.hbs +371 -0
- package/dist/core/llm/prompts/templates/v1/user.hbs +20 -0
- package/dist/core/llm/prompts/templates/v2/system.hbs +573 -0
- package/dist/core/llm/prompts/templates/v2/user.hbs +20 -0
- package/dist/core/llm/prompts/templates/v3/system.hbs +861 -0
- package/dist/core/llm/prompts/templates/v3/user.hbs +16 -0
- package/dist/core/llm/prompts/templates/v4/system.hbs +800 -0
- package/dist/core/llm/prompts/templates/v4/user.hbs +40 -0
- package/dist/core/llm/prompts/templates/v4.5/system.hbs +71 -0
- package/dist/core/llm/prompts/templates/v4.5/user.hbs +46 -0
- package/dist/core/llm/prompts/templates/v5/glossary/system.hbs +40 -0
- package/dist/core/llm/prompts/templates/v5/glossary/user.hbs +11 -0
- package/dist/core/llm/prompts/templates/v5/system.hbs +163 -0
- package/dist/core/llm/prompts/templates/v5/user.hbs +55 -0
- package/dist/core/pipeline/GroundingTransform.js +52 -0
- package/dist/core/pipeline/GroundingTransform.js.map +1 -0
- package/dist/core/pipeline/PipelineRunner.js +51 -0
- package/dist/core/pipeline/PipelineRunner.js.map +1 -0
- package/dist/core/pipeline/RelationFilterTransform.js +72 -0
- package/dist/core/pipeline/RelationFilterTransform.js.map +1 -0
- package/dist/core/pipeline/index.js +20 -0
- package/dist/core/pipeline/index.js.map +1 -0
- package/dist/core/processor/FileProcessor.js +184 -0
- package/dist/core/processor/FileProcessor.js.map +1 -0
- package/dist/core/processor/ProcessedRegistry.js +38 -0
- package/dist/core/processor/ProcessedRegistry.js.map +1 -0
- package/dist/core/processor/ast/AstSeedService.js +0 -0
- package/dist/core/processor/ast/AstSeedService.js.map +1 -0
- package/dist/core/processor/ast/AstSymbolStore.js +110 -0
- package/dist/core/processor/ast/AstSymbolStore.js.map +1 -0
- package/dist/core/processor/ast/index.js +19 -0
- package/dist/core/processor/ast/index.js.map +1 -0
- package/dist/core/processor/chunking/TextChunker.js +98 -0
- package/dist/core/processor/chunking/TextChunker.js.map +1 -0
- package/dist/core/processor/chunking/index.js +18 -0
- package/dist/core/processor/chunking/index.js.map +1 -0
- package/dist/core/processor/classifier/CONTENT_CLASSES.js +294 -0
- package/dist/core/processor/classifier/CONTENT_CLASSES.js.map +1 -0
- package/dist/core/processor/classifier/CascadeContentClassifier.js +107 -0
- package/dist/core/processor/classifier/CascadeContentClassifier.js.map +1 -0
- package/dist/core/processor/classifier/HeuristicContentClassifier.js +113 -0
- package/dist/core/processor/classifier/HeuristicContentClassifier.js.map +1 -0
- package/dist/core/processor/classifier/IContentTypeClassifier.js +3 -0
- package/dist/core/processor/classifier/IContentTypeClassifier.js.map +1 -0
- package/dist/core/processor/classifier/LlmContentClassifier.js +107 -0
- package/dist/core/processor/classifier/LlmContentClassifier.js.map +1 -0
- package/dist/core/processor/classifier/NER_DOMAIN_EXAMPLES.js +498 -0
- package/dist/core/processor/classifier/NER_DOMAIN_EXAMPLES.js.map +1 -0
- package/dist/core/processor/classifier/index.js +21 -0
- package/dist/core/processor/classifier/index.js.map +1 -0
- package/dist/core/processor/classifier/mergeClassifications.js +32 -0
- package/dist/core/processor/classifier/mergeClassifications.js.map +1 -0
- package/dist/core/processor/index.js +20 -0
- package/dist/core/processor/index.js.map +1 -0
- package/dist/core/processor/readers/AudioReader.js +462 -0
- package/dist/core/processor/readers/AudioReader.js.map +1 -0
- package/dist/core/processor/readers/BinaryReader.js +90 -0
- package/dist/core/processor/readers/BinaryReader.js.map +1 -0
- package/dist/core/processor/readers/ChandraPdfReader.js +187 -0
- package/dist/core/processor/readers/ChandraPdfReader.js.map +1 -0
- package/dist/core/processor/readers/ChatExportReader.js +365 -0
- package/dist/core/processor/readers/ChatExportReader.js.map +1 -0
- package/dist/core/processor/readers/DoclingReader.js +445 -0
- package/dist/core/processor/readers/DoclingReader.js.map +1 -0
- package/dist/core/processor/readers/EmailReader.js +259 -0
- package/dist/core/processor/readers/EmailReader.js.map +1 -0
- package/dist/core/processor/readers/EpubReader.js +175 -0
- package/dist/core/processor/readers/EpubReader.js.map +1 -0
- package/dist/core/processor/readers/FileReader.js +90 -0
- package/dist/core/processor/readers/FileReader.js.map +1 -0
- package/dist/core/processor/readers/FileReaderFactory.js +49 -0
- package/dist/core/processor/readers/FileReaderFactory.js.map +1 -0
- package/dist/core/processor/readers/HtmlReader.js +371 -0
- package/dist/core/processor/readers/HtmlReader.js.map +1 -0
- package/dist/core/processor/readers/ImageReader.js +162 -0
- package/dist/core/processor/readers/ImageReader.js.map +1 -0
- package/dist/core/processor/readers/JsonFileReader.js +232 -0
- package/dist/core/processor/readers/JsonFileReader.js.map +1 -0
- package/dist/core/processor/readers/JupyterReader.js +178 -0
- package/dist/core/processor/readers/JupyterReader.js.map +1 -0
- package/dist/core/processor/readers/LatexReader.js +176 -0
- package/dist/core/processor/readers/LatexReader.js.map +1 -0
- package/dist/core/processor/readers/MarkdownReader.js +289 -0
- package/dist/core/processor/readers/MarkdownReader.js.map +1 -0
- package/dist/core/processor/readers/MarkerPdfReader.js +193 -0
- package/dist/core/processor/readers/MarkerPdfReader.js.map +1 -0
- package/dist/core/processor/readers/MistralOcrReader.js +198 -0
- package/dist/core/processor/readers/MistralOcrReader.js.map +1 -0
- package/dist/core/processor/readers/OfficeReader.js +174 -0
- package/dist/core/processor/readers/OfficeReader.js.map +1 -0
- package/dist/core/processor/readers/PdfReader.js +116 -0
- package/dist/core/processor/readers/PdfReader.js.map +1 -0
- package/dist/core/processor/readers/RtfReader.js +107 -0
- package/dist/core/processor/readers/RtfReader.js.map +1 -0
- package/dist/core/processor/readers/SubtitleReader.js +145 -0
- package/dist/core/processor/readers/SubtitleReader.js.map +1 -0
- package/dist/core/processor/readers/TesseractPdfReader.js +183 -0
- package/dist/core/processor/readers/TesseractPdfReader.js.map +1 -0
- package/dist/core/processor/readers/TextReader.js +129 -0
- package/dist/core/processor/readers/TextReader.js.map +1 -0
- package/dist/core/processor/readers/TranscriptReader.js +234 -0
- package/dist/core/processor/readers/TranscriptReader.js.map +1 -0
- package/dist/core/processor/readers/image/imageMetadata.js +155 -0
- package/dist/core/processor/readers/image/imageMetadata.js.map +1 -0
- package/dist/core/processor/readers/index.js +41 -0
- package/dist/core/processor/readers/index.js.map +1 -0
- package/dist/core/processor/readers/referenceExtraction.js +198 -0
- package/dist/core/processor/readers/referenceExtraction.js.map +1 -0
- package/dist/core/processor/readers/stripReferences.js +59 -0
- package/dist/core/processor/readers/stripReferences.js.map +1 -0
- package/dist/core/processor/readers/transcript/turnPacking.js +81 -0
- package/dist/core/processor/readers/transcript/turnPacking.js.map +1 -0
- package/dist/core/progress/NdjsonProgressEmitter.js +30 -0
- package/dist/core/progress/NdjsonProgressEmitter.js.map +1 -0
- package/dist/core/progress/NoopProgressEmitter.js +15 -0
- package/dist/core/progress/NoopProgressEmitter.js.map +1 -0
- package/dist/core/progress/index.js +19 -0
- package/dist/core/progress/index.js.map +1 -0
- package/dist/core/trace/TraceWriter.js +100 -0
- package/dist/core/trace/TraceWriter.js.map +1 -0
- package/dist/core/trace/events.js +13 -0
- package/dist/core/trace/events.js.map +1 -0
- package/dist/core/trace/index.js +20 -0
- package/dist/core/trace/index.js.map +1 -0
- package/dist/core/trace/lineage.js +97 -0
- package/dist/core/trace/lineage.js.map +1 -0
- package/dist/evaluation/BenchmarkRunner.js +171 -0
- package/dist/evaluation/BenchmarkRunner.js.map +1 -0
- package/dist/evaluation/classifier/ClassifierAccuracy.js +185 -0
- package/dist/evaluation/classifier/ClassifierAccuracy.js.map +1 -0
- package/dist/evaluation/classifier/labeledSamples.js +379 -0
- package/dist/evaluation/classifier/labeledSamples.js.map +1 -0
- package/dist/evaluation/compare/goldCompare.js +126 -0
- package/dist/evaluation/compare/goldCompare.js.map +1 -0
- package/dist/evaluation/crossre/compareScoring.js +30 -0
- package/dist/evaluation/crossre/compareScoring.js.map +1 -0
- package/dist/evaluation/datasets/CrossREDataset.js +170 -0
- package/dist/evaluation/datasets/CrossREDataset.js.map +1 -0
- package/dist/evaluation/datasets/IDataset.js +3 -0
- package/dist/evaluation/datasets/IDataset.js.map +1 -0
- package/dist/evaluation/datasets/RebelDataset.js +117 -0
- package/dist/evaluation/datasets/RebelDataset.js.map +1 -0
- package/dist/evaluation/datasets/RedocredDataset.js +218 -0
- package/dist/evaluation/datasets/RedocredDataset.js.map +1 -0
- package/dist/evaluation/datasets/SemEval2010Dataset.js +150 -0
- package/dist/evaluation/datasets/SemEval2010Dataset.js.map +1 -0
- package/dist/evaluation/index.js +33 -0
- package/dist/evaluation/index.js.map +1 -0
- package/dist/evaluation/matching/ExactMatcher.js +75 -0
- package/dist/evaluation/matching/ExactMatcher.js.map +1 -0
- package/dist/evaluation/matching/SemanticMatcher.js +143 -0
- package/dist/evaluation/matching/SemanticMatcher.js.map +1 -0
- package/dist/evaluation/metrics/TripleMetrics.js +64 -0
- package/dist/evaluation/metrics/TripleMetrics.js.map +1 -0
- package/dist/evaluation/mine/MineCheckpoint.js +114 -0
- package/dist/evaluation/mine/MineCheckpoint.js.map +1 -0
- package/dist/evaluation/mine/MineDataset.js +208 -0
- package/dist/evaluation/mine/MineDataset.js.map +1 -0
- package/dist/evaluation/mine/MineReporter.js +98 -0
- package/dist/evaluation/mine/MineReporter.js.map +1 -0
- package/dist/evaluation/mine/MineRunner.js +148 -0
- package/dist/evaluation/mine/MineRunner.js.map +1 -0
- package/dist/evaluation/mine/MineScorer.js +127 -0
- package/dist/evaluation/mine/MineScorer.js.map +1 -0
- package/dist/evaluation/mine/types.js +12 -0
- package/dist/evaluation/mine/types.js.map +1 -0
- package/dist/evaluation/reporters/ConsoleReporter.js +55 -0
- package/dist/evaluation/reporters/ConsoleReporter.js.map +1 -0
- package/dist/evaluation/reporters/JsonReporter.js +50 -0
- package/dist/evaluation/reporters/JsonReporter.js.map +1 -0
- package/dist/index.js +28 -0
- package/dist/index.js.map +1 -0
- package/dist/quality/CompositeScore.js +61 -0
- package/dist/quality/CompositeScore.js.map +1 -0
- package/dist/quality/ConsistencyMetrics.js +70 -0
- package/dist/quality/ConsistencyMetrics.js.map +1 -0
- package/dist/quality/FactualMetrics.js +76 -0
- package/dist/quality/FactualMetrics.js.map +1 -0
- package/dist/quality/GraphHealthMetrics.js +68 -0
- package/dist/quality/GraphHealthMetrics.js.map +1 -0
- package/dist/quality/SemanticMetrics.js +102 -0
- package/dist/quality/SemanticMetrics.js.map +1 -0
- package/dist/quality/StructuralMetrics.js +60 -0
- package/dist/quality/StructuralMetrics.js.map +1 -0
- package/dist/quality/index.js +23 -0
- package/dist/quality/index.js.map +1 -0
- package/dist/shared/index.js +20 -0
- package/dist/shared/index.js.map +1 -0
- package/dist/shared/logger/Logger.js +3 -0
- package/dist/shared/logger/Logger.js.map +1 -0
- package/dist/shared/logger/LoggerFactory.js +75 -0
- package/dist/shared/logger/LoggerFactory.js.map +1 -0
- package/dist/shared/logger/index.js +19 -0
- package/dist/shared/logger/index.js.map +1 -0
- package/dist/shared/shutdown.js +30 -0
- package/dist/shared/shutdown.js.map +1 -0
- package/dist/shared/utils/agglomerativeCluster.js +269 -0
- package/dist/shared/utils/agglomerativeCluster.js.map +1 -0
- package/dist/shared/utils/astSymbols.js +69 -0
- package/dist/shared/utils/astSymbols.js.map +1 -0
- package/dist/shared/utils/cosineSimilarity.js +18 -0
- package/dist/shared/utils/cosineSimilarity.js.map +1 -0
- package/dist/shared/utils/directoryTree.js +184 -0
- package/dist/shared/utils/directoryTree.js.map +1 -0
- package/dist/shared/utils/documentOutline.js +74 -0
- package/dist/shared/utils/documentOutline.js.map +1 -0
- package/dist/shared/utils/index.js +24 -0
- package/dist/shared/utils/index.js.map +1 -0
- package/dist/shared/utils/jaroWinklerSimilarity.js +60 -0
- package/dist/shared/utils/jaroWinklerSimilarity.js.map +1 -0
- package/dist/shared/utils/parseJsonLenient.js +27 -0
- package/dist/shared/utils/parseJsonLenient.js.map +1 -0
- package/dist/shared/utils/readConfig.js +42 -0
- package/dist/shared/utils/readConfig.js.map +1 -0
- package/dist/shared/utils/readRtf.js +216 -0
- package/dist/shared/utils/readRtf.js.map +1 -0
- package/dist/shared/utils/softmax.js +26 -0
- package/dist/shared/utils/softmax.js.map +1 -0
- package/dist/types/ContentClass.js +3 -0
- package/dist/types/ContentClass.js.map +1 -0
- package/dist/types/CorpusProfile.js +3 -0
- package/dist/types/CorpusProfile.js.map +1 -0
- package/dist/types/IContradictionChecker.js +3 -0
- package/dist/types/IContradictionChecker.js.map +1 -0
- package/dist/types/ICorpusAnalyzer.js +3 -0
- package/dist/types/ICorpusAnalyzer.js.map +1 -0
- package/dist/types/IDirectoryProcessor.js +3 -0
- package/dist/types/IDirectoryProcessor.js.map +1 -0
- package/dist/types/IEmbeddingProvider.js +3 -0
- package/dist/types/IEmbeddingProvider.js.map +1 -0
- package/dist/types/IEmbeddingService.js +6 -0
- package/dist/types/IEmbeddingService.js.map +1 -0
- package/dist/types/IFileProcessor.js +3 -0
- package/dist/types/IFileProcessor.js.map +1 -0
- package/dist/types/IGroundingChecker.js +3 -0
- package/dist/types/IGroundingChecker.js.map +1 -0
- package/dist/types/IKnowledgeGraphBuilder.js +3 -0
- package/dist/types/IKnowledgeGraphBuilder.js.map +1 -0
- package/dist/types/IKnowledgeGraphExporter.js +3 -0
- package/dist/types/IKnowledgeGraphExporter.js.map +1 -0
- package/dist/types/IKnowledgeGraphMerger.js +3 -0
- package/dist/types/IKnowledgeGraphMerger.js.map +1 -0
- package/dist/types/IKnowledgeGraphSearch.js +3 -0
- package/dist/types/IKnowledgeGraphSearch.js.map +1 -0
- package/dist/types/ILLMProvider.js +3 -0
- package/dist/types/ILLMProvider.js.map +1 -0
- package/dist/types/ILLMService.js +3 -0
- package/dist/types/ILLMService.js.map +1 -0
- package/dist/types/IObjectDetector.js +3 -0
- package/dist/types/IObjectDetector.js.map +1 -0
- package/dist/types/IProcessingService.js +3 -0
- package/dist/types/IProcessingService.js.map +1 -0
- package/dist/types/IProgressEmitter.js +3 -0
- package/dist/types/IProgressEmitter.js.map +1 -0
- package/dist/types/IPromptManager.js +3 -0
- package/dist/types/IPromptManager.js.map +1 -0
- package/dist/types/KnowledgeGraph.js +3 -0
- package/dist/types/KnowledgeGraph.js.map +1 -0
- package/dist/types/MCPKnowledgeGraph.js +3 -0
- package/dist/types/MCPKnowledgeGraph.js.map +1 -0
- package/dist/types/Observation.js +21 -0
- package/dist/types/Observation.js.map +1 -0
- package/dist/types/ProcessingOptions.js +3 -0
- package/dist/types/ProcessingOptions.js.map +1 -0
- package/dist/types/index.js +40 -0
- package/dist/types/index.js.map +1 -0
- package/package.json +122 -0
|
@@ -0,0 +1,170 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
|
|
3
|
+
if (k2 === undefined) k2 = k;
|
|
4
|
+
var desc = Object.getOwnPropertyDescriptor(m, k);
|
|
5
|
+
if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
|
|
6
|
+
desc = { enumerable: true, get: function() { return m[k]; } };
|
|
7
|
+
}
|
|
8
|
+
Object.defineProperty(o, k2, desc);
|
|
9
|
+
}) : (function(o, m, k, k2) {
|
|
10
|
+
if (k2 === undefined) k2 = k;
|
|
11
|
+
o[k2] = m[k];
|
|
12
|
+
}));
|
|
13
|
+
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
|
|
14
|
+
Object.defineProperty(o, "default", { enumerable: true, value: v });
|
|
15
|
+
}) : function(o, v) {
|
|
16
|
+
o["default"] = v;
|
|
17
|
+
});
|
|
18
|
+
var __importStar = (this && this.__importStar) || (function () {
|
|
19
|
+
var ownKeys = function(o) {
|
|
20
|
+
ownKeys = Object.getOwnPropertyNames || function (o) {
|
|
21
|
+
var ar = [];
|
|
22
|
+
for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
|
|
23
|
+
return ar;
|
|
24
|
+
};
|
|
25
|
+
return ownKeys(o);
|
|
26
|
+
};
|
|
27
|
+
return function (mod) {
|
|
28
|
+
if (mod && mod.__esModule) return mod;
|
|
29
|
+
var result = {};
|
|
30
|
+
if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
|
|
31
|
+
__setModuleDefault(result, mod);
|
|
32
|
+
return result;
|
|
33
|
+
};
|
|
34
|
+
})();
|
|
35
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
36
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
37
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
38
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
39
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
40
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
41
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
42
|
+
});
|
|
43
|
+
};
|
|
44
|
+
var __asyncValues = (this && this.__asyncValues) || function (o) {
|
|
45
|
+
if (!Symbol.asyncIterator) throw new TypeError("Symbol.asyncIterator is not defined.");
|
|
46
|
+
var m = o[Symbol.asyncIterator], i;
|
|
47
|
+
return m ? m.call(o) : (o = typeof __values === "function" ? __values(o) : o[Symbol.iterator](), i = {}, verb("next"), verb("throw"), verb("return"), i[Symbol.asyncIterator] = function () { return this; }, i);
|
|
48
|
+
function verb(n) { i[n] = o[n] && function (v) { return new Promise(function (resolve, reject) { v = o[n](v), settle(resolve, reject, v.done, v.value); }); }; }
|
|
49
|
+
function settle(resolve, reject, d, v) { Promise.resolve(v).then(function(v) { resolve({ value: v, done: d }); }, reject); }
|
|
50
|
+
};
|
|
51
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
52
|
+
exports.CrossREDataset = void 0;
|
|
53
|
+
const fs = __importStar(require("fs"));
|
|
54
|
+
const path = __importStar(require("path"));
|
|
55
|
+
const readline = __importStar(require("readline"));
|
|
56
|
+
class CrossREDataset {
|
|
57
|
+
/**
|
|
58
|
+
* Load CrossRE samples.
|
|
59
|
+
*
|
|
60
|
+
* @param dataPath Path to a single domain JSON file OR a directory
|
|
61
|
+
* containing multiple `*-test.json` / `*-dev.json` files.
|
|
62
|
+
* When a directory is given every `.json` file inside is loaded.
|
|
63
|
+
* @param limit Maximum total samples to return (applied across all files).
|
|
64
|
+
* @param domain Optional domain filter. Accepts a single domain string
|
|
65
|
+
* (e.g. `"ai"`) or a comma-separated list (e.g. `"ai,news,science"`).
|
|
66
|
+
* When omitted all domains are included.
|
|
67
|
+
*/
|
|
68
|
+
load(dataPath, limit, domain) {
|
|
69
|
+
return __awaiter(this, void 0, void 0, function* () {
|
|
70
|
+
if (!fs.existsSync(dataPath)) {
|
|
71
|
+
throw new Error(`CrossRE data path not found: ${dataPath}\n` +
|
|
72
|
+
`Pass a single domain file (e.g. ./data/crossre/crossre_data/ai-test.json)\n` +
|
|
73
|
+
`or the crossre_data/ directory to load all domain splits.\n` +
|
|
74
|
+
`Files are available at: https://github.com/mainlp/CrossRE/tree/main/crossre_data`);
|
|
75
|
+
}
|
|
76
|
+
// Normalise domain filter: "ai,news,science" → Set { "ai", "news", "science" }
|
|
77
|
+
const domainFilter = domain
|
|
78
|
+
? new Set(domain.split(',').map(d => d.trim()).filter(Boolean))
|
|
79
|
+
: null;
|
|
80
|
+
const files = this.resolveFiles(dataPath);
|
|
81
|
+
const samples = [];
|
|
82
|
+
for (const file of files) {
|
|
83
|
+
if (samples.length >= limit)
|
|
84
|
+
break;
|
|
85
|
+
yield this.loadFile(file, limit - samples.length, domainFilter, samples);
|
|
86
|
+
}
|
|
87
|
+
return samples;
|
|
88
|
+
});
|
|
89
|
+
}
|
|
90
|
+
// ─── Private ───────────────────────────────────────────────────────────────
|
|
91
|
+
/** Return a sorted list of JSON files to load. */
|
|
92
|
+
resolveFiles(dataPath) {
|
|
93
|
+
const stat = fs.statSync(dataPath);
|
|
94
|
+
if (stat.isDirectory()) {
|
|
95
|
+
return fs.readdirSync(dataPath)
|
|
96
|
+
.filter(f => f.endsWith('.json'))
|
|
97
|
+
.sort()
|
|
98
|
+
.map(f => path.join(dataPath, f));
|
|
99
|
+
}
|
|
100
|
+
return [dataPath];
|
|
101
|
+
}
|
|
102
|
+
loadFile(filePath, limit, domainFilter, out) {
|
|
103
|
+
return __awaiter(this, void 0, void 0, function* () {
|
|
104
|
+
var _a, e_1, _b, _c;
|
|
105
|
+
const stream = fs.createReadStream(filePath, { encoding: 'utf-8' });
|
|
106
|
+
const rl = readline.createInterface({ input: stream, crlfDelay: Infinity });
|
|
107
|
+
try {
|
|
108
|
+
for (var _d = true, rl_1 = __asyncValues(rl), rl_1_1; rl_1_1 = yield rl_1.next(), _a = rl_1_1.done, !_a; _d = true) {
|
|
109
|
+
_c = rl_1_1.value;
|
|
110
|
+
_d = false;
|
|
111
|
+
const line = _c;
|
|
112
|
+
if (out.length >= limit)
|
|
113
|
+
break;
|
|
114
|
+
const trimmed = line.trim();
|
|
115
|
+
if (!trimmed)
|
|
116
|
+
continue;
|
|
117
|
+
let raw;
|
|
118
|
+
try {
|
|
119
|
+
raw = JSON.parse(trimmed);
|
|
120
|
+
}
|
|
121
|
+
catch (_e) {
|
|
122
|
+
continue;
|
|
123
|
+
}
|
|
124
|
+
const tokens = raw.sentence || [];
|
|
125
|
+
const nerTuples = raw.ner || [];
|
|
126
|
+
const relTuples = raw.relations || [];
|
|
127
|
+
if (tokens.length === 0 || nerTuples.length === 0 || relTuples.length === 0)
|
|
128
|
+
continue;
|
|
129
|
+
// Extract domain from doc_key prefix (e.g. "news-test-1" -> "news")
|
|
130
|
+
const detectedDomain = raw.doc_key.split('-')[0];
|
|
131
|
+
// Filter by domain if specified
|
|
132
|
+
if (domainFilter && !domainFilter.has(detectedDomain))
|
|
133
|
+
continue;
|
|
134
|
+
// Build span→mention lookup: "start,end" -> mention string
|
|
135
|
+
const spanToMention = new Map();
|
|
136
|
+
for (const [start, end] of nerTuples) {
|
|
137
|
+
const mention = tokens.slice(start, end + 1).join(' ');
|
|
138
|
+
spanToMention.set(`${start},${end}`, mention);
|
|
139
|
+
}
|
|
140
|
+
// Each relation tuple: [start1, end1, start2, end2, rel-type, ...]
|
|
141
|
+
const groundTruth = [];
|
|
142
|
+
for (const [s1, e1, s2, e2, relType] of relTuples) {
|
|
143
|
+
const subject = spanToMention.get(`${s1},${e1}`);
|
|
144
|
+
const object = spanToMention.get(`${s2},${e2}`);
|
|
145
|
+
if (subject && object) {
|
|
146
|
+
groundTruth.push({ subject, predicate: relType, object });
|
|
147
|
+
}
|
|
148
|
+
}
|
|
149
|
+
if (groundTruth.length === 0)
|
|
150
|
+
continue;
|
|
151
|
+
out.push({
|
|
152
|
+
id: raw.doc_key,
|
|
153
|
+
text: tokens.join(' '),
|
|
154
|
+
groundTruth,
|
|
155
|
+
domain: detectedDomain,
|
|
156
|
+
});
|
|
157
|
+
}
|
|
158
|
+
}
|
|
159
|
+
catch (e_1_1) { e_1 = { error: e_1_1 }; }
|
|
160
|
+
finally {
|
|
161
|
+
try {
|
|
162
|
+
if (!_d && !_a && (_b = rl_1.return)) yield _b.call(rl_1);
|
|
163
|
+
}
|
|
164
|
+
finally { if (e_1) throw e_1.error; }
|
|
165
|
+
}
|
|
166
|
+
});
|
|
167
|
+
}
|
|
168
|
+
}
|
|
169
|
+
exports.CrossREDataset = CrossREDataset;
|
|
170
|
+
//# sourceMappingURL=CrossREDataset.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"CrossREDataset.js","sourceRoot":"","sources":["../../../src/evaluation/datasets/CrossREDataset.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,uCAAyB;AACzB,2CAA6B;AAC7B,mDAAqC;AAiBrC,MAAa,cAAc;IACzB;;;;;;;;;;OAUG;IACG,IAAI,CAAC,QAAgB,EAAE,KAAa,EAAE,MAAe;;YACzD,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC,QAAQ,CAAC,EAAE,CAAC;gBAC7B,MAAM,IAAI,KAAK,CACb,gCAAgC,QAAQ,IAAI;oBAC5C,6EAA6E;oBAC7E,6DAA6D;oBAC7D,kFAAkF,CACnF,CAAC;YACJ,CAAC;YAED,+EAA+E;YAC/E,MAAM,YAAY,GAAuB,MAAM;gBAC7C,CAAC,CAAC,IAAI,GAAG,CAAC,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;gBAC/D,CAAC,CAAC,IAAI,CAAC;YAET,MAAM,KAAK,GAAG,IAAI,CAAC,YAAY,CAAC,QAAQ,CAAC,CAAC;YAC1C,MAAM,OAAO,GAAsB,EAAE,CAAC;YAEtC,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;gBACzB,IAAI,OAAO,CAAC,MAAM,IAAI,KAAK;oBAAE,MAAM;gBACnC,MAAM,IAAI,CAAC,QAAQ,CAAC,IAAI,EAAE,KAAK,GAAG,OAAO,CAAC,MAAM,EAAE,YAAY,EAAE,OAAO,CAAC,CAAC;YAC3E,CAAC;YAED,OAAO,OAAO,CAAC;QACjB,CAAC;KAAA;IAED,8EAA8E;IAE9E,kDAAkD;IAC1C,YAAY,CAAC,QAAgB;QACnC,MAAM,IAAI,GAAG,EAAE,CAAC,QAAQ,CAAC,QAAQ,CAAC,CAAC;QACnC,IAAI,IAAI,CAAC,WAAW,EAAE,EAAE,CAAC;YACvB,OAAO,EAAE,CAAC,WAAW,CAAC,QAAQ,CAAC;iBAC5B,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,QAAQ,CAAC,OAAO,CAAC,CAAC;iBAChC,IAAI,EAAE;iBACN,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC,CAAC;QACtC,CAAC;QACD,OAAO,CAAC,QAAQ,CAAC,CAAC;IACpB,CAAC;IAEa,QAAQ,CACpB,QAAgB,EAChB,KAAa,EACb,YAAgC,EAChC,GAAsB;;;YAEtB,MAAM,MAAM,GAAG,EAAE,CAAC,gBAAgB,CAAC,QAAQ,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,CAAC,CAAC;YACpE,MAAM,EAAE,GAAG,QAAQ,CAAC,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAC;;gBAE5E,KAAyB,eAAA,OAAA,cAAA,EAAE,CAAA,QAAA,gEAAE,CAAC;oBAAL,kBAAE;oBAAF,WAAE;oBAAhB,MAAM,IAAI,KAAA,CAAA;oBACnB,IAAI,GAAG,CAAC,MAAM,IAAI,KAAK;wBAAE,MAAM;oBAC/B,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC;oBAC5B,IAAI,CAAC,OAAO;wBAAE,SAAS;oBAEvB,IAAI,GAAmB,CAAC;oBACxB,IAAI,CAAC;wBACH,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;oBAC5B,CAAC;oBAAC,WAAM,CAAC;wBACP,SAAS;oBACX,CAAC;oBAED,MAAM,MAAM,GAAM,GAAG,CAAC,QAAQ,IAAI,EAAE,CAAC;oBACrC,MAAM,SAAS,GAAG,GAAG,CAAC,GAAG,IAAI,EAAE,CAAC;oBAChC,MAAM,SAAS,GAAG,GAAG,CAAC,SAAS,IAAI,EAAE,CAAC;oBAEtC,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,IAAI,SAAS,CAAC,MAAM,KAAK,CAAC,IAAI,SAAS,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAS;oBAEtF,oEAAoE;oBACpE,MAAM,cAAc,GAAG,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;oBAEjD,gCAAgC;oBAChC,IAAI,YAAY,IAAI,CAAC,YAAY,CAAC,GAAG,CAAC,cAAc,CAAC;wBAAE,SAAS;oBAEhE,2DAA2D;oBAC3D,MAAM,aAAa,GAAG,IAAI,GAAG,EAAkB,CAAC;oBAChD,KAAK,MAAM,CAAC,KAAK,EAAE,GAAG,CAAC,IAAI,SAAS,EAAE,CAAC;wBACrC,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;wBACvD,aAAa,CAAC,GAAG,CAAC,GAAG,KAAK,IAAI,GAAG,EAAE,EAAE,OAAO,CAAC,CAAC;oBAChD,CAAC;oBAED,mEAAmE;oBACnE,MAAM,WAAW,GAAc,EAAE,CAAC;oBAClC,KAAK,MAAM,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,OAAO,CAAC,IAAI,SAAS,EAAE,CAAC;wBAClD,MAAM,OAAO,GAAG,aAAa,CAAC,GAAG,CAAC,GAAG,EAAE,IAAI,EAAE,EAAE,CAAC,CAAC;wBACjD,MAAM,MAAM,GAAI,aAAa,CAAC,GAAG,CAAC,GAAG,EAAE,IAAI,EAAE,EAAE,CAAC,CAAC;wBACjD,IAAI,OAAO,IAAI,MAAM,EAAE,CAAC;4BACtB,WAAW,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,SAAS,EAAE,OAAO,EAAE,MAAM,EAAE,CAAC,CAAC;wBAC5D,CAAC;oBACH,CAAC;oBAED,IAAI,WAAW,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAS;oBAEvC,GAAG,CAAC,IAAI,CAAC;wBACP,EAAE,EAAE,GAAG,CAAC,OAAO;wBACf,IAAI,EAAE,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC;wBACtB,WAAW;wBACX,MAAM,EAAE,cAAc;qBACvB,CAAC,CAAC;gBACL,CAAC;;;;;;;;;QACH,CAAC;KAAA;CACF;AAhHD,wCAgHC"}
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"IDataset.js","sourceRoot":"","sources":["../../../src/evaluation/datasets/IDataset.ts"],"names":[],"mappings":""}
|
|
@@ -0,0 +1,117 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
|
|
3
|
+
if (k2 === undefined) k2 = k;
|
|
4
|
+
var desc = Object.getOwnPropertyDescriptor(m, k);
|
|
5
|
+
if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
|
|
6
|
+
desc = { enumerable: true, get: function() { return m[k]; } };
|
|
7
|
+
}
|
|
8
|
+
Object.defineProperty(o, k2, desc);
|
|
9
|
+
}) : (function(o, m, k, k2) {
|
|
10
|
+
if (k2 === undefined) k2 = k;
|
|
11
|
+
o[k2] = m[k];
|
|
12
|
+
}));
|
|
13
|
+
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
|
|
14
|
+
Object.defineProperty(o, "default", { enumerable: true, value: v });
|
|
15
|
+
}) : function(o, v) {
|
|
16
|
+
o["default"] = v;
|
|
17
|
+
});
|
|
18
|
+
var __importStar = (this && this.__importStar) || (function () {
|
|
19
|
+
var ownKeys = function(o) {
|
|
20
|
+
ownKeys = Object.getOwnPropertyNames || function (o) {
|
|
21
|
+
var ar = [];
|
|
22
|
+
for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
|
|
23
|
+
return ar;
|
|
24
|
+
};
|
|
25
|
+
return ownKeys(o);
|
|
26
|
+
};
|
|
27
|
+
return function (mod) {
|
|
28
|
+
if (mod && mod.__esModule) return mod;
|
|
29
|
+
var result = {};
|
|
30
|
+
if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
|
|
31
|
+
__setModuleDefault(result, mod);
|
|
32
|
+
return result;
|
|
33
|
+
};
|
|
34
|
+
})();
|
|
35
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
36
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
37
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
38
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
39
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
40
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
41
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
42
|
+
});
|
|
43
|
+
};
|
|
44
|
+
var __asyncValues = (this && this.__asyncValues) || function (o) {
|
|
45
|
+
if (!Symbol.asyncIterator) throw new TypeError("Symbol.asyncIterator is not defined.");
|
|
46
|
+
var m = o[Symbol.asyncIterator], i;
|
|
47
|
+
return m ? m.call(o) : (o = typeof __values === "function" ? __values(o) : o[Symbol.iterator](), i = {}, verb("next"), verb("throw"), verb("return"), i[Symbol.asyncIterator] = function () { return this; }, i);
|
|
48
|
+
function verb(n) { i[n] = o[n] && function (v) { return new Promise(function (resolve, reject) { v = o[n](v), settle(resolve, reject, v.done, v.value); }); }; }
|
|
49
|
+
function settle(resolve, reject, d, v) { Promise.resolve(v).then(function(v) { resolve({ value: v, done: d }); }, reject); }
|
|
50
|
+
};
|
|
51
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
52
|
+
exports.RebelDataset = void 0;
|
|
53
|
+
const fs = __importStar(require("fs"));
|
|
54
|
+
const readline = __importStar(require("readline"));
|
|
55
|
+
class RebelDataset {
|
|
56
|
+
load(dataPath, limit) {
|
|
57
|
+
return __awaiter(this, void 0, void 0, function* () {
|
|
58
|
+
var _a, e_1, _b, _c;
|
|
59
|
+
var _d, _e, _f, _g, _h, _j;
|
|
60
|
+
if (!fs.existsSync(dataPath)) {
|
|
61
|
+
throw new Error(`REBEL dataset not found at: ${dataPath}\n` +
|
|
62
|
+
`Download from: https://huggingface.co/datasets/Babelscape/rebel-dataset`);
|
|
63
|
+
}
|
|
64
|
+
const samples = [];
|
|
65
|
+
const stream = fs.createReadStream(dataPath, { encoding: 'utf-8' });
|
|
66
|
+
const rl = readline.createInterface({ input: stream, crlfDelay: Infinity });
|
|
67
|
+
try {
|
|
68
|
+
for (var _k = true, rl_1 = __asyncValues(rl), rl_1_1; rl_1_1 = yield rl_1.next(), _a = rl_1_1.done, !_a; _k = true) {
|
|
69
|
+
_c = rl_1_1.value;
|
|
70
|
+
_k = false;
|
|
71
|
+
const line = _c;
|
|
72
|
+
if (samples.length >= limit)
|
|
73
|
+
break;
|
|
74
|
+
const trimmed = line.trim();
|
|
75
|
+
if (!trimmed)
|
|
76
|
+
continue;
|
|
77
|
+
let raw;
|
|
78
|
+
try {
|
|
79
|
+
raw = JSON.parse(trimmed);
|
|
80
|
+
}
|
|
81
|
+
catch (_l) {
|
|
82
|
+
continue;
|
|
83
|
+
}
|
|
84
|
+
const text = raw.text || '';
|
|
85
|
+
if (!text || !raw.triples || raw.triples.length === 0)
|
|
86
|
+
continue;
|
|
87
|
+
const groundTruth = [];
|
|
88
|
+
for (const t of raw.triples) {
|
|
89
|
+
const subject = (_e = (_d = t.subject) === null || _d === void 0 ? void 0 : _d.surfaceform) === null || _e === void 0 ? void 0 : _e.trim();
|
|
90
|
+
const predicate = (_g = (_f = t.predicate) === null || _f === void 0 ? void 0 : _f.surfaceform) === null || _g === void 0 ? void 0 : _g.trim();
|
|
91
|
+
const object = (_j = (_h = t.object) === null || _h === void 0 ? void 0 : _h.surfaceform) === null || _j === void 0 ? void 0 : _j.trim();
|
|
92
|
+
if (subject && predicate && object) {
|
|
93
|
+
groundTruth.push({ subject, predicate, object });
|
|
94
|
+
}
|
|
95
|
+
}
|
|
96
|
+
if (groundTruth.length === 0)
|
|
97
|
+
continue;
|
|
98
|
+
samples.push({
|
|
99
|
+
id: raw.docid || String(samples.length),
|
|
100
|
+
text,
|
|
101
|
+
groundTruth,
|
|
102
|
+
});
|
|
103
|
+
}
|
|
104
|
+
}
|
|
105
|
+
catch (e_1_1) { e_1 = { error: e_1_1 }; }
|
|
106
|
+
finally {
|
|
107
|
+
try {
|
|
108
|
+
if (!_k && !_a && (_b = rl_1.return)) yield _b.call(rl_1);
|
|
109
|
+
}
|
|
110
|
+
finally { if (e_1) throw e_1.error; }
|
|
111
|
+
}
|
|
112
|
+
return samples;
|
|
113
|
+
});
|
|
114
|
+
}
|
|
115
|
+
}
|
|
116
|
+
exports.RebelDataset = RebelDataset;
|
|
117
|
+
//# sourceMappingURL=RebelDataset.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"RebelDataset.js","sourceRoot":"","sources":["../../../src/evaluation/datasets/RebelDataset.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,uCAAyB;AACzB,mDAAqC;AA6CrC,MAAa,YAAY;IACjB,IAAI,CAAC,QAAgB,EAAE,KAAa;;;;YACxC,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC,QAAQ,CAAC,EAAE,CAAC;gBAC7B,MAAM,IAAI,KAAK,CACb,+BAA+B,QAAQ,IAAI;oBAC3C,yEAAyE,CAC1E,CAAC;YACJ,CAAC;YAED,MAAM,OAAO,GAAsB,EAAE,CAAC;YACtC,MAAM,MAAM,GAAG,EAAE,CAAC,gBAAgB,CAAC,QAAQ,EAAE,EAAE,QAAQ,EAAE,OAAO,EAAE,CAAC,CAAC;YACpE,MAAM,EAAE,GAAG,QAAQ,CAAC,eAAe,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,QAAQ,EAAE,CAAC,CAAC;;gBAE5E,KAAyB,eAAA,OAAA,cAAA,EAAE,CAAA,QAAA,gEAAE,CAAC;oBAAL,kBAAE;oBAAF,WAAE;oBAAhB,MAAM,IAAI,KAAA,CAAA;oBACnB,IAAI,OAAO,CAAC,MAAM,IAAI,KAAK;wBAAE,MAAM;oBACnC,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC;oBAC5B,IAAI,CAAC,OAAO;wBAAE,SAAS;oBAEvB,IAAI,GAAiB,CAAC;oBACtB,IAAI,CAAC;wBACH,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;oBAC5B,CAAC;oBAAC,WAAM,CAAC;wBACP,SAAS;oBACX,CAAC;oBAED,MAAM,IAAI,GAAG,GAAG,CAAC,IAAI,IAAI,EAAE,CAAC;oBAC5B,IAAI,CAAC,IAAI,IAAI,CAAC,GAAG,CAAC,OAAO,IAAI,GAAG,CAAC,OAAO,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAS;oBAEhE,MAAM,WAAW,GAAc,EAAE,CAAC;oBAClC,KAAK,MAAM,CAAC,IAAI,GAAG,CAAC,OAAO,EAAE,CAAC;wBAC5B,MAAM,OAAO,GAAK,MAAA,MAAA,CAAC,CAAC,OAAO,0CAAE,WAAW,0CAAE,IAAI,EAAE,CAAC;wBACjD,MAAM,SAAS,GAAG,MAAA,MAAA,CAAC,CAAC,SAAS,0CAAE,WAAW,0CAAE,IAAI,EAAE,CAAC;wBACnD,MAAM,MAAM,GAAM,MAAA,MAAA,CAAC,CAAC,MAAM,0CAAE,WAAW,0CAAE,IAAI,EAAE,CAAC;wBAChD,IAAI,OAAO,IAAI,SAAS,IAAI,MAAM,EAAE,CAAC;4BACnC,WAAW,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,SAAS,EAAE,MAAM,EAAE,CAAC,CAAC;wBACnD,CAAC;oBACH,CAAC;oBAED,IAAI,WAAW,CAAC,MAAM,KAAK,CAAC;wBAAE,SAAS;oBAEvC,OAAO,CAAC,IAAI,CAAC;wBACX,EAAE,EAAE,GAAG,CAAC,KAAK,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,CAAC;wBACvC,IAAI;wBACJ,WAAW;qBACZ,CAAC,CAAC;gBACL,CAAC;;;;;;;;;YAED,OAAO,OAAO,CAAC;QACjB,CAAC;KAAA;CACF;AAjDD,oCAiDC"}
|
|
@@ -0,0 +1,218 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
|
|
3
|
+
if (k2 === undefined) k2 = k;
|
|
4
|
+
var desc = Object.getOwnPropertyDescriptor(m, k);
|
|
5
|
+
if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
|
|
6
|
+
desc = { enumerable: true, get: function() { return m[k]; } };
|
|
7
|
+
}
|
|
8
|
+
Object.defineProperty(o, k2, desc);
|
|
9
|
+
}) : (function(o, m, k, k2) {
|
|
10
|
+
if (k2 === undefined) k2 = k;
|
|
11
|
+
o[k2] = m[k];
|
|
12
|
+
}));
|
|
13
|
+
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
|
|
14
|
+
Object.defineProperty(o, "default", { enumerable: true, value: v });
|
|
15
|
+
}) : function(o, v) {
|
|
16
|
+
o["default"] = v;
|
|
17
|
+
});
|
|
18
|
+
var __importStar = (this && this.__importStar) || (function () {
|
|
19
|
+
var ownKeys = function(o) {
|
|
20
|
+
ownKeys = Object.getOwnPropertyNames || function (o) {
|
|
21
|
+
var ar = [];
|
|
22
|
+
for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
|
|
23
|
+
return ar;
|
|
24
|
+
};
|
|
25
|
+
return ownKeys(o);
|
|
26
|
+
};
|
|
27
|
+
return function (mod) {
|
|
28
|
+
if (mod && mod.__esModule) return mod;
|
|
29
|
+
var result = {};
|
|
30
|
+
if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
|
|
31
|
+
__setModuleDefault(result, mod);
|
|
32
|
+
return result;
|
|
33
|
+
};
|
|
34
|
+
})();
|
|
35
|
+
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
|
|
36
|
+
function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
|
|
37
|
+
return new (P || (P = Promise))(function (resolve, reject) {
|
|
38
|
+
function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
|
|
39
|
+
function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
|
|
40
|
+
function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
|
|
41
|
+
step((generator = generator.apply(thisArg, _arguments || [])).next());
|
|
42
|
+
});
|
|
43
|
+
};
|
|
44
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
45
|
+
exports.RedocredDataset = void 0;
|
|
46
|
+
const fs = __importStar(require("fs"));
|
|
47
|
+
// RE-DocRED JSON format (Qin et al., 2022):
|
|
48
|
+
// Array of documents, each with:
|
|
49
|
+
// {
|
|
50
|
+
// title: "...",
|
|
51
|
+
// sents: [["The", "Loud", "Tour", ...], [...], ...], // sentences as token arrays
|
|
52
|
+
// vertexSet: [ // entity groups
|
|
53
|
+
// [{ name, pos, sent_id, type, global_pos, index }, ...], // multiple mentions
|
|
54
|
+
// ...
|
|
55
|
+
// ],
|
|
56
|
+
// labels: [{ r: "P577", h: 0, t: 6, evidence: [1] }, ...] // relations (Wikidata PIDs)
|
|
57
|
+
// }
|
|
58
|
+
//
|
|
59
|
+
// Entity name resolution: longest mention in the group (most specific surface form).
|
|
60
|
+
// Text reconstruction: token-join all sentences with a single space between sentences.
|
|
61
|
+
// Relation labels: Wikidata property IDs are mapped to human-readable names via
|
|
62
|
+
// WIKIDATA_PROP_LABELS so the semantic matcher can compare them meaningfully.
|
|
63
|
+
// ─── Wikidata property ID → readable label ────────────────────────────────────
|
|
64
|
+
// Covers all 96 RE-DocRED relation types.
|
|
65
|
+
const WIKIDATA_PROP_LABELS = {
|
|
66
|
+
P6: 'head of government',
|
|
67
|
+
P17: 'country',
|
|
68
|
+
P19: 'place of birth',
|
|
69
|
+
P20: 'place of death',
|
|
70
|
+
P22: 'father',
|
|
71
|
+
P25: 'mother',
|
|
72
|
+
P26: 'spouse',
|
|
73
|
+
P27: 'country of citizenship',
|
|
74
|
+
P30: 'continent',
|
|
75
|
+
P31: 'instance of',
|
|
76
|
+
P35: 'head of state',
|
|
77
|
+
P36: 'capital',
|
|
78
|
+
P37: 'official language',
|
|
79
|
+
P39: 'position held',
|
|
80
|
+
P40: 'child',
|
|
81
|
+
P50: 'author',
|
|
82
|
+
P54: 'member of sports team',
|
|
83
|
+
P57: 'director',
|
|
84
|
+
P58: 'screenwriter',
|
|
85
|
+
P69: 'educated at',
|
|
86
|
+
P86: 'composer',
|
|
87
|
+
P102: 'member of political party',
|
|
88
|
+
P108: 'employer',
|
|
89
|
+
P112: 'founded by',
|
|
90
|
+
P118: 'league',
|
|
91
|
+
P123: 'publisher',
|
|
92
|
+
P127: 'owned by',
|
|
93
|
+
P131: 'located in administrative entity',
|
|
94
|
+
P136: 'genre',
|
|
95
|
+
P137: 'operator',
|
|
96
|
+
P140: 'religion or worldview',
|
|
97
|
+
P150: 'contains administrative entity',
|
|
98
|
+
P155: 'follows',
|
|
99
|
+
P156: 'followed by',
|
|
100
|
+
P159: 'headquarters location',
|
|
101
|
+
P161: 'cast member',
|
|
102
|
+
P162: 'producer',
|
|
103
|
+
P166: 'award received',
|
|
104
|
+
P170: 'creator',
|
|
105
|
+
P171: 'parent taxon',
|
|
106
|
+
P172: 'ethnic group',
|
|
107
|
+
P175: 'performer',
|
|
108
|
+
P176: 'manufacturer',
|
|
109
|
+
P178: 'developer',
|
|
110
|
+
P179: 'series',
|
|
111
|
+
P190: 'twinned administrative body',
|
|
112
|
+
P194: 'legislative body',
|
|
113
|
+
P205: 'basin country',
|
|
114
|
+
P206: 'located next to body of water',
|
|
115
|
+
P241: 'military branch',
|
|
116
|
+
P264: 'record label',
|
|
117
|
+
P272: 'production company',
|
|
118
|
+
P276: 'location',
|
|
119
|
+
P279: 'subclass of',
|
|
120
|
+
P355: 'subsidiary',
|
|
121
|
+
P361: 'part of',
|
|
122
|
+
P364: 'original language',
|
|
123
|
+
P400: 'platform',
|
|
124
|
+
P403: 'mouth of watercourse',
|
|
125
|
+
P449: 'original network',
|
|
126
|
+
P463: 'member of',
|
|
127
|
+
P488: 'chairperson',
|
|
128
|
+
P495: 'country of origin',
|
|
129
|
+
P527: 'has part',
|
|
130
|
+
P551: 'residence',
|
|
131
|
+
P569: 'date of birth',
|
|
132
|
+
P570: 'date of death',
|
|
133
|
+
P571: 'inception date',
|
|
134
|
+
P576: 'dissolved date',
|
|
135
|
+
P577: 'publication date',
|
|
136
|
+
P580: 'start time',
|
|
137
|
+
P582: 'end time',
|
|
138
|
+
P585: 'point in time',
|
|
139
|
+
P607: 'conflict',
|
|
140
|
+
P674: 'characters',
|
|
141
|
+
P676: 'lyrics by',
|
|
142
|
+
P706: 'located on terrain feature',
|
|
143
|
+
P710: 'participant',
|
|
144
|
+
P737: 'influenced by',
|
|
145
|
+
P740: 'location of formation',
|
|
146
|
+
P749: 'parent organization',
|
|
147
|
+
P800: 'notable work',
|
|
148
|
+
P807: 'separated from',
|
|
149
|
+
P840: 'narrative location',
|
|
150
|
+
P937: 'work location',
|
|
151
|
+
P1001: 'applies to jurisdiction',
|
|
152
|
+
P1056: 'product or material produced',
|
|
153
|
+
P1198: 'unemployment rate',
|
|
154
|
+
P1336: 'territory claimed by',
|
|
155
|
+
P1344: 'participant in',
|
|
156
|
+
P1365: 'replaces',
|
|
157
|
+
P1366: 'replaced by',
|
|
158
|
+
P1376: 'capital of',
|
|
159
|
+
P1412: 'languages spoken or written',
|
|
160
|
+
P1441: 'present in work',
|
|
161
|
+
P3373: 'sibling',
|
|
162
|
+
};
|
|
163
|
+
// ─── Dataset loader ───────────────────────────────────────────────────────────
|
|
164
|
+
class RedocredDataset {
|
|
165
|
+
/**
|
|
166
|
+
* Load RE-DocRED samples.
|
|
167
|
+
*
|
|
168
|
+
* @param dataPath Path to one of the JSON files:
|
|
169
|
+
* data/redocred/test_revised.json ← use for benchmarking
|
|
170
|
+
* data/redocred/dev_revised.json
|
|
171
|
+
* data/redocred/train_revised.json
|
|
172
|
+
* @param limit Maximum samples to return (0 / MAX_SAFE_INTEGER = all).
|
|
173
|
+
* @param domain Unused — RE-DocRED has no domain splits. Accepted for
|
|
174
|
+
* interface compatibility; pass undefined.
|
|
175
|
+
*/
|
|
176
|
+
load(dataPath, limit, _domain) {
|
|
177
|
+
return __awaiter(this, void 0, void 0, function* () {
|
|
178
|
+
var _a, _b, _c, _d;
|
|
179
|
+
if (!fs.existsSync(dataPath)) {
|
|
180
|
+
throw new Error(`RE-DocRED dataset not found at: ${dataPath}\n` +
|
|
181
|
+
`Download from: https://github.com/tonytan48/Re-DocRED\n` +
|
|
182
|
+
`Expected files: test_revised.json, dev_revised.json, train_revised.json`);
|
|
183
|
+
}
|
|
184
|
+
const raw = JSON.parse(fs.readFileSync(dataPath, 'utf-8'));
|
|
185
|
+
const samples = [];
|
|
186
|
+
for (let i = 0; i < raw.length && samples.length < limit; i++) {
|
|
187
|
+
const doc = raw[i];
|
|
188
|
+
if (!((_a = doc.sents) === null || _a === void 0 ? void 0 : _a.length) || !((_b = doc.vertexSet) === null || _b === void 0 ? void 0 : _b.length) || !((_c = doc.labels) === null || _c === void 0 ? void 0 : _c.length))
|
|
189
|
+
continue;
|
|
190
|
+
// Reconstruct full text by joining tokens across all sentences
|
|
191
|
+
const text = doc.sents.map(sent => sent.join(' ')).join(' ');
|
|
192
|
+
// Build entity index: entity group idx → canonical name (longest mention)
|
|
193
|
+
const entityNames = doc.vertexSet.map(mentions => mentions.reduce((best, m) => m.name.length > best.length ? m.name : best, ''));
|
|
194
|
+
// Build ground-truth triplets
|
|
195
|
+
const groundTruth = [];
|
|
196
|
+
for (const label of doc.labels) {
|
|
197
|
+
const subject = entityNames[label.h];
|
|
198
|
+
const object = entityNames[label.t];
|
|
199
|
+
const predicate = (_d = WIKIDATA_PROP_LABELS[label.r]) !== null && _d !== void 0 ? _d : label.r;
|
|
200
|
+
if (subject && object && predicate) {
|
|
201
|
+
groundTruth.push({ subject, predicate, object });
|
|
202
|
+
}
|
|
203
|
+
}
|
|
204
|
+
if (groundTruth.length === 0)
|
|
205
|
+
continue;
|
|
206
|
+
samples.push({
|
|
207
|
+
id: `${doc.title.replace(/\s+/g, '_')}_${i}`,
|
|
208
|
+
text,
|
|
209
|
+
groundTruth,
|
|
210
|
+
// No domain field in RE-DocRED; omit rather than set to undefined
|
|
211
|
+
});
|
|
212
|
+
}
|
|
213
|
+
return samples;
|
|
214
|
+
});
|
|
215
|
+
}
|
|
216
|
+
}
|
|
217
|
+
exports.RedocredDataset = RedocredDataset;
|
|
218
|
+
//# sourceMappingURL=RedocredDataset.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"RedocredDataset.js","sourceRoot":"","sources":["../../../src/evaluation/datasets/RedocredDataset.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,uCAAyB;AAGzB,4CAA4C;AAC5C,iCAAiC;AACjC,IAAI;AACJ,kBAAkB;AAClB,sFAAsF;AACtF,4EAA4E;AAC5E,oFAAoF;AACpF,UAAU;AACV,OAAO;AACP,2FAA2F;AAC3F,IAAI;AACJ,EAAE;AACF,qFAAqF;AACrF,uFAAuF;AACvF,gFAAgF;AAChF,gFAAgF;AAEhF,iFAAiF;AACjF,0CAA0C;AAC1C,MAAM,oBAAoB,GAA2B;IACnD,EAAE,EAAK,oBAAoB;IAC3B,GAAG,EAAI,SAAS;IAChB,GAAG,EAAI,gBAAgB;IACvB,GAAG,EAAI,gBAAgB;IACvB,GAAG,EAAI,QAAQ;IACf,GAAG,EAAI,QAAQ;IACf,GAAG,EAAI,QAAQ;IACf,GAAG,EAAI,wBAAwB;IAC/B,GAAG,EAAI,WAAW;IAClB,GAAG,EAAI,aAAa;IACpB,GAAG,EAAI,eAAe;IACtB,GAAG,EAAI,SAAS;IAChB,GAAG,EAAI,mBAAmB;IAC1B,GAAG,EAAI,eAAe;IACtB,GAAG,EAAI,OAAO;IACd,GAAG,EAAI,QAAQ;IACf,GAAG,EAAI,uBAAuB;IAC9B,GAAG,EAAI,UAAU;IACjB,GAAG,EAAI,cAAc;IACrB,GAAG,EAAI,aAAa;IACpB,GAAG,EAAI,UAAU;IACjB,IAAI,EAAG,2BAA2B;IAClC,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,YAAY;IACnB,IAAI,EAAG,QAAQ;IACf,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,kCAAkC;IACzC,IAAI,EAAG,OAAO;IACd,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,uBAAuB;IAC9B,IAAI,EAAG,gCAAgC;IACvC,IAAI,EAAG,SAAS;IAChB,IAAI,EAAG,aAAa;IACpB,IAAI,EAAG,uBAAuB;IAC9B,IAAI,EAAG,aAAa;IACpB,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,gBAAgB;IACvB,IAAI,EAAG,SAAS;IAChB,IAAI,EAAG,cAAc;IACrB,IAAI,EAAG,cAAc;IACrB,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,cAAc;IACrB,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,QAAQ;IACf,IAAI,EAAG,6BAA6B;IACpC,IAAI,EAAG,kBAAkB;IACzB,IAAI,EAAG,eAAe;IACtB,IAAI,EAAG,+BAA+B;IACtC,IAAI,EAAG,iBAAiB;IACxB,IAAI,EAAG,cAAc;IACrB,IAAI,EAAG,oBAAoB;IAC3B,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,aAAa;IACpB,IAAI,EAAG,YAAY;IACnB,IAAI,EAAG,SAAS;IAChB,IAAI,EAAG,mBAAmB;IAC1B,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,sBAAsB;IAC7B,IAAI,EAAG,kBAAkB;IACzB,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,aAAa;IACpB,IAAI,EAAG,mBAAmB;IAC1B,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,eAAe;IACtB,IAAI,EAAG,eAAe;IACtB,IAAI,EAAG,gBAAgB;IACvB,IAAI,EAAG,gBAAgB;IACvB,IAAI,EAAG,kBAAkB;IACzB,IAAI,EAAG,YAAY;IACnB,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,eAAe;IACtB,IAAI,EAAG,UAAU;IACjB,IAAI,EAAG,YAAY;IACnB,IAAI,EAAG,WAAW;IAClB,IAAI,EAAG,4BAA4B;IACnC,IAAI,EAAG,aAAa;IACpB,IAAI,EAAG,eAAe;IACtB,IAAI,EAAG,uBAAuB;IAC9B,IAAI,EAAG,qBAAqB;IAC5B,IAAI,EAAG,cAAc;IACrB,IAAI,EAAG,gBAAgB;IACvB,IAAI,EAAG,oBAAoB;IAC3B,IAAI,EAAG,eAAe;IACtB,KAAK,EAAE,yBAAyB;IAChC,KAAK,EAAE,8BAA8B;IACrC,KAAK,EAAE,mBAAmB;IAC1B,KAAK,EAAE,sBAAsB;IAC7B,KAAK,EAAE,gBAAgB;IACvB,KAAK,EAAE,UAAU;IACjB,KAAK,EAAE,aAAa;IACpB,KAAK,EAAE,YAAY;IACnB,KAAK,EAAE,6BAA6B;IACpC,KAAK,EAAE,iBAAiB;IACxB,KAAK,EAAE,SAAS;CACjB,CAAC;AA2BF,iFAAiF;AAEjF,MAAa,eAAe;IAC1B;;;;;;;;;;OAUG;IACG,IAAI,CAAC,QAAgB,EAAE,KAAa,EAAE,OAAgB;;;YAC1D,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC,QAAQ,CAAC,EAAE,CAAC;gBAC7B,MAAM,IAAI,KAAK,CACb,mCAAmC,QAAQ,IAAI;oBAC/C,yDAAyD;oBACzD,yEAAyE,CAC1E,CAAC;YACJ,CAAC;YAED,MAAM,GAAG,GAAkB,IAAI,CAAC,KAAK,CAAC,EAAE,CAAC,YAAY,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC,CAAC;YAC1E,MAAM,OAAO,GAAsB,EAAE,CAAC;YAEtC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,GAAG,CAAC,MAAM,IAAI,OAAO,CAAC,MAAM,GAAG,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC9D,MAAM,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,CAAC;gBAEnB,IAAI,CAAC,CAAA,MAAA,GAAG,CAAC,KAAK,0CAAE,MAAM,CAAA,IAAI,CAAC,CAAA,MAAA,GAAG,CAAC,SAAS,0CAAE,MAAM,CAAA,IAAI,CAAC,CAAA,MAAA,GAAG,CAAC,MAAM,0CAAE,MAAM,CAAA;oBAAE,SAAS;gBAElF,+DAA+D;gBAC/D,MAAM,IAAI,GAAG,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;gBAE7D,0EAA0E;gBAC1E,MAAM,WAAW,GAAG,GAAG,CAAC,SAAS,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAC/C,QAAQ,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,CAC9E,CAAC;gBAEF,8BAA8B;gBAC9B,MAAM,WAAW,GAAc,EAAE,CAAC;gBAClC,KAAK,MAAM,KAAK,IAAI,GAAG,CAAC,MAAM,EAAE,CAAC;oBAC/B,MAAM,OAAO,GAAK,WAAW,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;oBACvC,MAAM,MAAM,GAAM,WAAW,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;oBACvC,MAAM,SAAS,GAAG,MAAA,oBAAoB,CAAC,KAAK,CAAC,CAAC,CAAC,mCAAI,KAAK,CAAC,CAAC,CAAC;oBAE3D,IAAI,OAAO,IAAI,MAAM,IAAI,SAAS,EAAE,CAAC;wBACnC,WAAW,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,SAAS,EAAE,MAAM,EAAE,CAAC,CAAC;oBACnD,CAAC;gBACH,CAAC;gBAED,IAAI,WAAW,CAAC,MAAM,KAAK,CAAC;oBAAE,SAAS;gBAEvC,OAAO,CAAC,IAAI,CAAC;oBACX,EAAE,EAAW,GAAG,GAAG,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,IAAI,CAAC,EAAE;oBACrD,IAAI;oBACJ,WAAW;oBACX,kEAAkE;iBACnE,CAAC,CAAC;YACL,CAAC;YAED,OAAO,OAAO,CAAC;QACjB,CAAC;KAAA;CACF;AA7DD,0CA6DC"}
|