Fixing errors of repeated texts in OpenAI model

.gitignore

@@ -170,3 +170,4 @@ __marimo__/
 *.pdf
 *.csv
 cache/
+sample/

app.py

@@ -202,14 +202,15 @@ def create_graph(json_data, model_name=MODEL_LIST[0]):
     
     # Configure network display
     network.from_nx(G)
-    network.barnes_hut(
-        gravity=-3000,
-        central_gravity=0.3,
-        spring_length=50,
-        spring_strength=0.001,
-        damping=0.09,
-        overlap=0,
-    )
+    if model_name == MODEL_LIST[0]:
+        network.barnes_hut(
+            gravity=-3000,
+            central_gravity=0.3,
+            spring_length=50,
+            spring_strength=0.001,
+            damping=0.09,
+            overlap=0,
+        )
 
     # Customize node appearance
     for node in network.nodes:
@@ -253,6 +254,7 @@ def process_and_visualize(text, model_name, progress=gr.Progress()):
     is_first_example = text == EXAMPLES[0][0]
     
     # Ensure RAG is initialized
+    # TODO: Clear all the previous inserted texts
     asyncio.run(model.initialize_rag()) 
 
     # Try to load from cache if it's the first example

llm_graph.py

@@ -45,17 +45,18 @@ class LLMGraph:
         Initialize the LightRAG instance with the specified embedding dimension.
         """
 
-        if self.rag is None:
-            self.rag = LightRAG(
-                working_dir=WORKING_DIR,
-                llm_model_func=self._llm_model_func,
-                embedding_func=EmbeddingFunc(
-                    embedding_dim=embedding_dimension,
-                    max_token_size=8192,
-                    func=self._embedding_func,
-                ),
-            )
-        # TODO: Check if this works as expected
+        # if self.rag is None:
+        # TODO: Check how to clear all the previous inserted texts
+        self.rag = LightRAG(
+            working_dir=WORKING_DIR,
+            llm_model_func=self._llm_model_func,
+            embedding_func=EmbeddingFunc(
+                embedding_dim=embedding_dimension,
+                max_token_size=8192,
+                func=self._embedding_func,
+            ),
+        )
+
         await self.rag.initialize_storages()
         await initialize_pipeline_status()
 
@@ -171,7 +172,7 @@ class LLMGraph:
             os.makedirs(WORKING_DIR, exist_ok=True)
             
             # Use LightRAG with Azure OpenAI
-            # TODO: Clear all the previous inserted texts first
+            # TODO: Clear all the previous inserted texts
             self.rag.insert(text) # Insert the text into the RAG storage
 
             # Wait for GRAPHML_FILE to be created

sample/kv_store_doc_status.json

@@ -7,29 +7,13 @@
     ],
     "content_summary": "The family of Azerbaijan President Ilham Aliyev leads a charmed, glamorous life, thanks in part to financial interests in almost every sector of the economy. His wife, Mehriban, comes from the privileged and powerful Pashayev family that owns banks, ...",
     "content_length": 1074,
-    "created_at": "2025-08-31T15:50:59.506391+00:00",
-    "updated_at": "2025-08-31T15:52:26.018288+00:00",
+    "created_at": "2025-09-01T14:53:10.896398+00:00",
+    "updated_at": "2025-09-01T14:54:44.898862+00:00",
     "file_path": "unknown_source",
-    "track_id": "insert_20250831_235059_6946ff78",
+    "track_id": "insert_20250901_225310_a186b861",
     "metadata": {
-      "processing_start_time": 1756655459,
-      "processing_end_time": 1756655546
-    }
-  },
-  "doc-eea199eb7feea197ebb82e9333a2d2f2": {
-    "status": "processing",
-    "chunks_count": 1,
-    "chunks_list": [
-      "chunk-eea199eb7feea197ebb82e9333a2d2f2"
-    ],
-    "content_summary": "Les jardins du Luxembourg, situés au cœur du sixième arrondissement de Paris, offrent un véritable havre de paix aux citadins pressés. Créés au début du dix-septième siècle sur l'initiative de Marie de Médicis, ces jardins à la française s'étendent s...",
-    "content_length": 697,
-    "created_at": "2025-08-31T15:54:38.060638+00:00",
-    "updated_at": "2025-08-31T15:54:38.068349+00:00",
-    "file_path": "unknown_source",
-    "track_id": "insert_20250831_235438_22d326d7",
-    "metadata": {
-      "processing_start_time": 1756655678
+      "processing_start_time": 1756738390,
+      "processing_end_time": 1756738484
     }
   }
 }