Move extractors (microsoft#1516)

* Consolidate graph embedding and umap

* Consolidate claim extraction

* Consolidate graph extractor

* Move graph utils

* Move summarizers

* Semver

---------

Co-authored-by: Alonso Guevara <[email protected]>

Author: natoverse

.semversioner/next-release/patch-20241213181544864279.json

@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "Move extractor code to co-locate with operations."
+}

graphrag/config/models/claim_extraction_config.py

@@ -37,12 +37,7 @@ class ClaimExtractionConfig(LLMConfig):
 
     def resolved_strategy(self, root_dir: str, encoding_model: str | None) -> dict:
         """Get the resolved claim extraction strategy."""
-        from graphrag.index.operations.extract_covariates import (
-            ExtractClaimsStrategyType,
-        )
-
         return self.strategy or {
-            "type": ExtractClaimsStrategyType.graph_intelligence,
             "llm": self.llm.model_dump(),
             **self.parallelization.model_dump(),
             "extraction_prompt": (Path(root_dir) / self.prompt)

graphrag/config/models/embed_graph_config.py

@@ -36,7 +36,7 @@ class EmbedGraphConfig(BaseModel):
 
     def resolved_strategy(self) -> dict:
         """Get the resolved node2vec strategy."""
-        from graphrag.index.operations.embed_graph import (
+        from graphrag.index.operations.embed_graph.typing import (
             EmbedGraphStrategyType,
         )
 

graphrag/index/flows/create_final_community_reports.py

@@ -12,7 +12,12 @@
 )
 
 from graphrag.cache.pipeline_cache import PipelineCache
-from graphrag.index.graph.extractors.community_reports.schemas import (
+from graphrag.index.operations.summarize_communities import (
+    prepare_community_reports,
+    restore_community_hierarchy,
+    summarize_communities,
+)
+from graphrag.index.operations.summarize_communities.community_reports_extractor.schemas import (
     CLAIM_DESCRIPTION,
     CLAIM_DETAILS,
     CLAIM_ID,
@@ -32,11 +37,6 @@
     NODE_ID,
     NODE_NAME,
 )
-from graphrag.index.operations.summarize_communities import (
-    prepare_community_reports,
-    restore_community_hierarchy,
-    summarize_communities,
-)
 
 
 async def create_final_community_reports(

graphrag/index/flows/create_final_covariates.py

@@ -13,7 +13,7 @@
 )
 
 from graphrag.cache.pipeline_cache import PipelineCache
-from graphrag.index.operations.extract_covariates import (
+from graphrag.index.operations.extract_covariates.extract_covariates import (
     extract_covariates,
 )
 

graphrag/index/flows/create_final_nodes.py

@@ -11,8 +11,8 @@
 )
 
 from graphrag.index.operations.create_graph import create_graph
-from graphrag.index.operations.embed_graph import embed_graph
-from graphrag.index.operations.layout_graph import layout_graph
+from graphrag.index.operations.embed_graph.embed_graph import embed_graph
+from graphrag.index.operations.layout_graph.layout_graph import layout_graph
 
 
 def create_final_nodes(