Update main.py

main.py

@@ -5,43 +5,20 @@ from llama_index.core import StorageContext, load_index_from_storage, VectorStor
 from llama_index.llms.huggingface import HuggingFaceInferenceAPI  
 from llama_index.embeddings.huggingface import HuggingFaceEmbedding  
 from huggingface_hub import InferenceClient  
+from transformers import AutoTokenizer, AutoModel
+
 
 # Ensure HF_TOKEN is set  
 HF_TOKEN = os.getenv("HF_TOKEN")  
 if not HF_TOKEN:  
     raise ValueError("HF_TOKEN environment variable not set.")  
 
-repo_id = "FacebookAI/xlm-roberta-base"  
+repo_id = "meta-llama/Meta-Llama-3-8B-Instruct"  
 llm_client = InferenceClient(  
     model=repo_id,  
     token=HF_TOKEN,  
 )  
 
-# Configure Llama index settings  
-# Settings.llm = HuggingFaceInferenceAPI(  
-#     model_name=repo_id,  
-#     tokenizer_name=repo_id,  
-#     context_window=3000,  
-#     token=HF_TOKEN,  
-#     max_new_tokens=512,  
-#     generate_kwargs={"temperature": 0.1},  
-# )  
-# Settings.embed_model = HuggingFaceEmbedding(  
-#     model_name="BAAI/bge-small-en-v1.5"  
-# )  
-# # Configure Llama index settings  
-# Settings.llm = HuggingFaceInferenceAPI(  
-#     model_name="xlm-roberta-base",  
-#     tokenizer_name="xlm-roberta-base",  
-#     context_window=3000,  
-#     token=HF_TOKEN,  
-#     max_new_tokens=512,  
-#     generate_kwargs={"temperature": 0.1},  
-# )  
-
-# Settings.embed_model = HuggingFaceEmbedding(  
-#     model_name="sentence-transformers/paraphrase-xlm-r-100langs-v1"  
-# )
 # Configure Llama index settings  
 Settings.llm = HuggingFaceInferenceAPI(  
     model_name=repo_id,  
@@ -51,11 +28,18 @@ Settings.llm = HuggingFaceInferenceAPI(
     max_new_tokens=512,  
     generate_kwargs={"temperature": 0.1},  
 )  
-
-Settings.embed_model = HuggingFaceEmbedding(  
-    model_name="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2"  # Updated model name  
+# Settings.embed_model = HuggingFaceEmbedding(  
+#     model_name="BAAI/bge-small-en-v1.5"  
+# )  
+# Replace the embedding model with XLM-R
+Settings.embed_model = HuggingFaceEmbedding(
+    model_name="xlm-roberta-base"  # XLM-RoBERTa model for multilingual support
 )
 
+# Configure tokenizer and model if required
+tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-base")
+model = AutoModel.from_pretrained("xlm-roberta-base")
+
 PERSIST_DIR = "db"  
 PDF_DIRECTORY = 'data'  
 
@@ -83,42 +67,29 @@ def data_ingestion_from_directory():
     index.storage_context.persist(persist_dir=PERSIST_DIR)  
 
 def handle_query(query):  
-    # chat_text_qa_msgs = [  
-    #     (  
-    #         "user",  
-    #         """  
-    #         You are the Taj Hotel chatbot, known as Taj Hotel Helper. Your goal is to provide accurate and professional answers to user queries based on the information available about the Taj Hotel. Always respond clearly and concisely, ideally within 10-15 words. If you don't know the answer, say so politely.  
-    #         Context:  
-    #         {context_str}  
-    #         User's Question:  
-    #         {query_str}  
-    #         """  
-    #     )  
-    # ]  
-    chat_text_qa_msgs = [
-        (
-            "user",
-            """
-            You are the Taj Hotel chatbot, known as Taj Hotel Helper. 
-            Your goal is to provide accurate and professional answers to 
-            user queries about the Taj Hotel in the language they use: 
-            English, Telugu, or Hindi. Always respond clearly and concisely, 
-            ideally within 10-15 words. If you don't know the answer, say so politely.
+    chat_text_qa_msgs = [  
+        (  
+            "user",  
+            """  
+            You are the Taj Hotel chatbot, known as Taj Hotel Helper.   
+            Your goal is to provide accurate and professional answers to   
+            user queries about the Taj Hotel in the language they use:   
+            English, Telugu, or Hindi. Always respond clearly and concisely,   
+            ideally within 10-15 words. If you don't know the answer, say so politely.  
     
-            Context:
-            {context_str}
+            Context:  
+            {context_str}  
     
-            User's Question:
-            {query_str}
+            User's Question:  
+            {query_str}  
     
-            Language-Specific Guidance:
-            - For English: Respond in English.
-            - For Telugu: తెలుగు లో సమాధానం ఇవ్వండి.
-            - For Hindi: हिंदी में उत्तर दें.
-            """
-        )
+            Language-Specific Guidance:  
+            - For English: Respond in English.  
+            - For Telugu: తెలుగు లో సమాధానం ఇవ్వండి.  
+            - For Hindi: हिंदी में उत्तर दें.  
+            """  
+        )  
     ]
-
     
     text_qa_template = ChatPromptTemplate.from_messages(chat_text_qa_msgs)