refactor

app.py

@@ -17,6 +17,7 @@ from dotenv import load_dotenv
 from threading import Lock
 from gradio.routes import Request
 from qdrant_client import QdrantClient
+import json
 
 # TESTING DEBUG LOG
 from auditqa.logging_config import setup_logging
@@ -44,7 +45,7 @@ scheduler = CommitScheduler(
      repo_id=model_config.get('app','repo_id'),
      repo_type="dataset",
      folder_path=JSON_DATASET_DIR,
-     path_in_repo="audit_chatbot",
+     path_in_repo=model_config.get('app','path_in_repo'),
      token=SPACES_LOG,
      every=2) # TESTING: every 2 seconds
 
@@ -145,10 +146,6 @@ session_manager = SessionManager()
 
 async def chat(query, history, sources, reports, subtype, year, client_ip=None, session_id=None):
     """Update chat function to handle session data"""
-    # TESTING: DEBUG LOG
-    logger.debug(f"Chat function called with query: {query}")
-    logger.debug(f"Client IP: {client_ip}")
-    logger.debug(f"Session ID: {session_id}")
     
     if not session_id: # Session managment
         session_id = session_manager.create_session(client_ip)
@@ -160,7 +157,7 @@ async def chat(query, history, sources, reports, subtype, year, client_ip=None,
     # Get session data
     session_data = session_manager.get_session_data(session_id)
     session_duration = session_manager.get_session_duration(session_id)
-    logger.debug(f"Session duration: {session_duration}")
+    logger.debug(f"Session duration calculated: {session_duration}")
     
     print(f">> NEW QUESTION : {query}")
     print(f"history:{history}")
@@ -220,22 +217,22 @@ async def chat(query, history, sources, reports, subtype, year, client_ip=None,
     timestamp = str(datetime.now().timestamp())
     logs_data = {
         "session_id": session_id,
-        "client_ip": client_ip,
         "client_location": session_data['location_info'],
         "session_duration_seconds": session_duration,
-        # "system_prompt": SYSTEM_PROMPT,
-        # "sources": sources,
-        # "reports": reports,
-        # "subtype": subtype,
+        # "system_prompt": SYSTEM_PROMPT, #REMOVED FOR TESTING
+        # "sources": sources, #REMOVED FOR TESTING
+        # "reports": reports, #REMOVED FOR TESTING
+        # "subtype": subtype, #REMOVED FOR TESTING
         "year": year,
         "question": query,
         "retriever": model_config.get('retriever','MODEL'),
         "endpoint_type": model_config.get('reader','TYPE'),
         "reader": model_config.get('reader','NVIDIA_MODEL'),
-        # "docs": [doc.page_content for doc in context_retrieved],
+        # "docs": [doc.page_content for doc in context_retrieved], #REMOVED FOR TESTING
         "answer": "",
         "time": timestamp,
     }
+    logger.debug(f"Logs data before save: {json.dumps(logs_data, indent=2)}")
 
     if model_config.get('reader','TYPE') == 'NVIDIA':
         chat_model = nvidia_client()
@@ -303,8 +300,9 @@ async def chat(query, history, sources, reports, subtype, year, client_ip=None,
     try:
         # Save log after streaming is complete
         save_logs(scheduler, JSON_DATASET_PATH, logs_data)
+        logger.debug(f"Logs saved successfully")
     except Exception as e:
-        logging.error(e)
+        logger.error(f"Error saving logs: {str(e)}")
 
 
 

model_params.cfg

@@ -13,4 +13,5 @@ NVIDIA_MODEL = meta-llama/Llama-3.1-8B-Instruct
 NVIDIA_ENDPOINT = https://huggingface.co/api/integrations/dgx/v1
 MAX_TOKENS = 512
 [app]
-repo_id = mtyrrell/audit_chatbot
+repo_id = mtyrrell/audit_chatbot #GIZ/spaces_log
+path_in_repo = . # audit_chatbot