Added predictions using local LLM

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 from typing import List
 import json
 from Classes.Owiki_Class import OWiki
+from Classes.run_local_LLM import LocalLLM
 
 def predict(query :str , chat_history : List[tuple[str,str]] , invocation_type : str = "OIC" , schemas : dict= {}) -> str:
     with open("config.json",'r') as f:
@@ -11,7 +12,8 @@ def predict(query :str , chat_history : List[tuple[str,str]] , invocation_type :
         try:
             res = a.create_sql_agent(query,schemas)
         except Exception as e:
-            return "Due to limited compute, I am unable to answer at this moment. Please upgrade your deployment space."  
+            res = local_llm.predict(query, invocation_type, schemas)
+            return res or "Due to limited compute, I am unable to answer at this moment. Please upgrade your deployment space."  
     elif invocation_type == "OIC":
         try:
             chat = ""
@@ -19,7 +21,8 @@ def predict(query :str , chat_history : List[tuple[str,str]] , invocation_type :
                 chat+= f"User: {user} Bot: {bot}\n\n"
             res = a.search_from_db(query, chat)
         except Exception as e:
-            return "Due to limited compute, I am unable to answer at this moment. Please upgrade your deployment space."  
+            res = local_llm.predict(query, invocation_type, schemas)
+            return res or "Due to limited compute, I am unable to answer at this moment. Please upgrade your deployment space."  
     return res