Store descriptive metrics identified by pipeline run id

Author: alekszievr

cognee/__init__.py

@@ -8,6 +8,7 @@
 from cognee.modules.visualization.cognee_network_visualization import (
     cognee_network_visualization,
 )
+from .modules.data.operations.store_descriptive_metrics import store_descriptive_metrics
 
 # Pipelines
 from .modules import pipelines

cognee/modules/data/methods/__init__.py

@@ -11,5 +11,3 @@
 # Delete
 from .delete_dataset import delete_dataset
 from .delete_data import delete_data
-
-from .store_descriptive_metrics import store_descriptive_metrics

cognee/modules/data/methods/store_descriptive_metrics.py

@@ -0,0 +1,62 @@
+from cognee.infrastructure.engine import DataPoint
+from cognee.infrastructure.databases.relational import get_relational_engine
+from sqlalchemy import select
+from sqlalchemy.sql import func
+from cognee.modules.data.models import Data
+from cognee.modules.data.models import GraphMetrics
+import uuid
+from cognee.infrastructure.databases.graph import get_graph_engine
+from cognee.modules.pipelines.models import PipelineRun
+
+
+async def fetch_token_count(db_engine) -> int:
+    """
+    Fetches and sums token counts from the database.
+
+    Returns:
+        int: The total number of tokens across all documents.
+    """
+
+    async with db_engine.get_async_session() as session:
+        token_count_sum = await session.execute(select(func.sum(Data.token_count)))
+        token_count_sum = token_count_sum.scalar()
+
+    return token_count_sum
+
+
+async def store_descriptive_metrics(pipeline_runs: list[PipelineRun], include_optional: bool):
+    db_engine = get_relational_engine()
+    graph_engine = await get_graph_engine()
+
+    metrics_for_pipeline_runs = []
+
+    async with db_engine.get_async_session() as session:
+        for pipeline_run in pipeline_runs:
+            existing_metrics = await session.execute(
+                select(GraphMetrics).where(GraphMetrics.id == pipeline_run.pipeline_run_id)
+            )
+            existing_metrics = existing_metrics.scalars().first()
+
+            if existing_metrics:
+                metrics_for_pipeline_runs.append(existing_metrics)
+            else:
+                graph_metrics = await graph_engine.get_graph_metrics(include_optional)
+                metrics = GraphMetrics(
+                    id=pipeline_run.pipeline_run_id,
+                    num_tokens=await fetch_token_count(db_engine),
+                    num_nodes=graph_metrics["num_nodes"],
+                    num_edges=graph_metrics["num_edges"],
+                    mean_degree=graph_metrics["mean_degree"],
+                    edge_density=graph_metrics["edge_density"],
+                    num_connected_components=graph_metrics["num_connected_components"],
+                    sizes_of_connected_components=graph_metrics["sizes_of_connected_components"],
+                    num_selfloops=graph_metrics["num_selfloops"],
+                    diameter=graph_metrics["diameter"],
+                    avg_shortest_path_length=graph_metrics["avg_shortest_path_length"],
+                    avg_clustering=graph_metrics["avg_clustering"],
+                )
+                metrics_for_pipeline_runs.append(metrics)
+                session.add(metrics)
+        await session.commit()
+
+    return metrics_for_pipeline_runs

cognee/modules/data/operations/store_descriptive_metrics.py

@@ -180,10 +180,15 @@ async def main(enable_steps):
 
     # Step 3: Create knowledge graph
     if enable_steps.get("cognify"):
-        await cognee.cognify()
+        pipeline_runs = await cognee.cognify()
         print("Knowledge graph created.")
 
-    # Step 4: Query insights
+    # Step 4: Calculate descriptive metrics
+    if enable_steps.get("graph_metrics"):
+        await cognee.store_descriptive_metrics(pipeline_runs, include_optional=True)
+        print("Descriptive graph metrics saved to database.")
+
+    # Step 5: Query insights
     if enable_steps.get("retriever"):
         search_results = await cognee.search(
             query_type=SearchType.GRAPH_COMPLETION, query_text="Who has experience in design tools?"
@@ -201,6 +206,7 @@ async def main(enable_steps):
         "prune_system": rebuild_kg,
         "add_text": rebuild_kg,
         "cognify": rebuild_kg,
+        "graph_metrics": rebuild_kg,
         "retriever": retrieve,
     }