Update rag.py

rag.py

@@ -43,8 +43,8 @@ except Exception as e:
     print(f"Error loading dataset: {e}")
     dataset = []
 
-# Precompute embeddings
-dataset_questions = [item.get("Question", "").lower().strip() for item in dataset]
+# Precompute normalized dataset embeddings
+dataset_questions = [normalize_input(item.get("Question", "")) for item in dataset]
 dataset_answers = [item.get("Answer", "") for item in dataset]
 dataset_embeddings = similarity_model.encode(dataset_questions, convert_to_tensor=True)
 
@@ -82,7 +82,7 @@ def query_groq_llm(prompt, model_name="llama3-70b-8192"):
         print(f"Error querying Groq API: {e}")
         return ""
 
-# Main logic function to be called from Gradio
+# Main logic function to be called from Gradio or elsewhere
 def get_best_answer(user_input):
     if not user_input.strip():
         return "Please enter a valid question."
@@ -106,7 +106,7 @@ def get_best_answer(user_input):
             "🔗 https://ue.edu.pk/allfeestructure.php"
         )
 
-    # Normalize only for similarity
+    # Normalize input for similarity
     normalized_input = normalize_input(user_input_lower)
     user_embedding = similarity_model.encode(normalized_input, convert_to_tensor=True)
     similarities = util.pytorch_cos_sim(user_embedding, dataset_embeddings)[0]