Update rag_functions.py

rag_functions.py

@@ -118,7 +118,7 @@ def retrieve_documents(query, index, doc_chunks, k=5):
     Returns:
         DataFrame: Top-k relevant document chunks, including the original context and link.
     """
-    query_embedding = encode_documents(pd.DataFrame({'chunk': [query]}))[0] # Encode our query text
+    query_embedding = encode_query(pd.DataFrame({'chunk': [query]}))[0] # Encode our query text
     query_embedding_np = query_embedding.squeeze().numpy().reshape(1, -1) # Transform our query text to a numpy array
     D, I = index.search(query_embedding_np, k) # Search for embeddings similar to our query
     return doc_chunks.iloc[I[0]] # Return the similar embeddings