Update app.py

app.py

@@ -17,7 +17,7 @@ from langchain_community.retrievers import PineconeHybridSearchRetriever
 from langchain_groq import ChatGroq
 
 # Load environment variables
-# load_dotenv(".env")
+load_dotenv(".env")
 USER_AGENT = os.getenv("USER_AGENT")
 GROQ_API_KEY = os.getenv("GROQ_API_KEY")
 SECRET_KEY = os.getenv("SECRET_KEY")
@@ -47,10 +47,36 @@ def initialize_pinecone(index_name: str):
         print(f"Error initializing Pinecone: {e}")
         raise
 
+
+##################################################
+##          Change down here
+##################################################
+
 # Initialize Pinecone index and BM25 encoder
 pinecone_index = initialize_pinecone("traveler-demo-website-vectorstore")
 bm25 = BM25Encoder().load("./bm25_traveler_website.json")
 
+### This is for UAE Legislation Website
+# pinecone_index = initialize_pinecone("uae-legislation-site-data")
+# bm25 = BM25Encoder().load("./bm25_uae_legislation_data.json")
+
+
+### This is for u.ae Website
+# pinecone_index = initialize_pinecone("vector-store-index")
+# bm25 = BM25Encoder().load("./bm25_u.ae.json")
+
+
+# #### This is for UAE Economic Department Website
+# pinecone_index = initialize_pinecone("uae-department-of-economics-site-data")
+# bm25 = BM25Encoder().load("./bm25_uae_department_of_economics_data.json")
+
+
+
+##################################################
+##################################################
+
+old_embed_model = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
+
 # Initialize models and retriever
 embed_model = HuggingFaceEmbeddings(model_name="Alibaba-NLP/gte-large-en-v1.5", model_kwargs={"trust_remote_code":True})
 retriever = PineconeHybridSearchRetriever(
@@ -62,7 +88,7 @@ retriever = PineconeHybridSearchRetriever(
 )
 
 # Initialize LLM
-llm = ChatGroq(model="llama-3.1-70b-versatile", temperature=0, max_tokens=1024, max_retries=2)
+llm = ChatGroq(model="llama-3.1-8b-instant", temperature=0, max_tokens=1024, max_retries=2)
 
 # Contextualization prompt and retriever
 contextualize_q_system_prompt = """Given a chat history and the latest user question \
@@ -82,6 +108,7 @@ history_aware_retriever = create_history_aware_retriever(llm, retriever, context
 # QA system prompt and chain
 qa_system_prompt = """You are a highly skilled information retrieval assistant. Use the following context to answer questions effectively. \
 If you don't know the answer, simply state that you don't know. \
+Your answer should be in {language} language. \
 Provide answers in proper HTML format and keep them concise. \
 
 When responding to queries, follow these guidelines: \
@@ -96,6 +123,8 @@ When responding to queries, follow these guidelines: \
     
     3. Formatting for Readability: \
         - The answer should be in a proper HTML format with appropriate tags. \
+        - For arabic language response align the text to right and convert numbers also.
+        - Double check if the language of answer is correct or not.
         - Use bullet points or numbered lists where applicable to present information clearly. \
         - Highlight key details using bold or italics. \
         - Provide proper and meaningful abbreviations for urls. Do not include naked urls. \
@@ -134,6 +163,7 @@ conversational_rag_chain = RunnableWithMessageHistory(
     get_session_history,
     input_messages_key="input",
     history_messages_key="chat_history",
+    language_message_key="language",
     output_messages_key="answer",
 )
 
@@ -153,12 +183,17 @@ def handle_disconnect():
 @socketio.on('message')
 def handle_message(data):
     question = data.get('question')
+    language = data.get('language')
+    if "en" in language:
+        language = "English"
+    else:
+        language = "Arabic"
     session_id = data.get('session_id', SESSION_ID_DEFAULT)
     chain = conversational_rag_chain.pick("answer")
 
     try:
         for chunk in chain.stream(
-                {"input": question},
+                {"input": question, 'language': language},
                 config={"configurable": {"session_id": session_id}},
             ):
             emit('response', chunk, room=request.sid)
@@ -174,4 +209,5 @@ def index_view():
 
 # Main function to run the app
 if __name__ == '__main__': 
-    socketio.run(app, debug=False)
+    print("Hello world")
+    socketio.run(app, debug=True)