Merge branch 'main' into config-refactor

microsoft · Jan 9, 2025 · 7744fac · 7744fac
2 parents f77f730 + 37fd7a7
commit 7744fac
Show file tree

Hide file tree

Showing 7 changed files with 31 additions and 8 deletions.
diff --git a/.semversioner/1.1.1.json b/.semversioner/1.1.1.json
@@ -0,0 +1,14 @@
+{
+  "changes": [
+    {
+      "description": "Fix a bug on creating community hierarchy for dynamic search",
+      "type": "patch"
+    },
+    {
+      "description": "Increase LOCAL_SEARCH_COMMUNITY_PROP to 15%",
+      "type": "patch"
+    }
+  ],
+  "created_at": "2025-01-08T21:53:16+00:00",
+  "version": "1.1.1"
+}
diff --git a/.semversioner/next-release/minor-20250109193320338859.json b/.semversioner/next-release/minor-20250109193320338859.json
@@ -0,0 +1,4 @@
+{
+  "type": "minor",
+  "description": "small bug fix"
+}
diff --git a/.semversioner/next-release/patch-20250107205718575866.json b/.semversioner/next-release/patch-20250107205718575866.json
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,6 +1,11 @@
 # Changelog
 Note: version releases in the 0.x.y range may introduce breaking changes.
 
+## 1.1.1
+
+- patch: Fix a bug on creating community hierarchy for dynamic search
+- patch: Increase LOCAL_SEARCH_COMMUNITY_PROP to 15%
+
 ## 1.1.0
 
 - minor: Make gleanings independent of encoding

diff --git a/graphrag/cli/query.py b/graphrag/cli/query.py
@@ -276,7 +276,7 @@ def run_basic_search(
     dataframe_dict = _resolve_output_files(
         config=config,
         output_list=[
-            "create_final_text_units.parquet",
+            "create_final_text_units",
         ],
     )
     final_text_units: pd.DataFrame = dataframe_dict["create_final_text_units"]

diff --git a/graphrag/query/context_builder/dynamic_community_selection.py b/graphrag/query/context_builder/dynamic_community_selection.py
@@ -58,19 +58,22 @@ def __init__(
         self.reports = {report.community_id: report for report in community_reports}
         # mapping from community to sub communities
         self.node2children = {
-            community.id: (
+            community.short_id: (
                 []
                 if community.sub_community_ids is None
-                else community.sub_community_ids
+                else [str(x) for x in community.sub_community_ids]
             )
             for community in communities
+            if community.short_id is not None
         }
+
         # mapping from community to parent community
         self.node2parent: dict[str, str] = {
             sub_community: community
             for community, sub_communities in self.node2children.items()
             for sub_community in sub_communities
         }
+
         # mapping from level to communities
         self.levels: dict[str, list[str]] = {}
 
@@ -182,5 +185,6 @@ async def select(self, query: str) -> tuple[list[CommunityReport], dict[str, Any
             llm_info["prompt_tokens"],
             llm_info["output_tokens"],
         )
+
         llm_info["ratings"] = ratings
         return community_reports, llm_info
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,7 +1,7 @@
 [tool.poetry]
 name = "graphrag"
 # Maintainers: do not change the version here manually, use ./scripts/release.sh
-version = "1.1.0"
+version = "1.1.1"
 description = "GraphRAG: A graph-based retrieval-augmented generation (RAG) system."
 authors = [
     "Alonso Guevara Fernández <[email protected]>",