Update rag.py

rag.py

@@ -8,6 +8,7 @@ from PIL import Image, ImageDraw, ImageFont
 import numpy as np
 from dotenv import load_dotenv
 import os
+import pandas as pd  # <-- Required for Excel logging
 
 # Load environment variables
 load_dotenv()
@@ -18,7 +19,7 @@ groq_client = Groq(api_key=os.getenv("GROQ_API_KEY"))
 # Load models and dataset
 similarity_model = SentenceTransformer('paraphrase-MiniLM-L6-v2')
 
-# Load dataset (automatically using the path)
+# Load dataset
 with open('dataset.json', 'r') as f:
     dataset = json.load(f)
 
@@ -46,7 +47,7 @@ def query_groq_llm(prompt, model_name="llama3-70b-8192"):
 def get_best_answer(user_input):
     user_input_lower = user_input.lower().strip()
 
-    # 👉 Check if question is about fee
+    # 👉 Fee-specific shortcut
     if any(keyword in user_input_lower for keyword in ["fee", "fees", "charges", "semester fee"]):
         return (
             "💰 For complete and up-to-date fee details for this program, we recommend visiting the official University of Education fee structure page.\n"
@@ -54,12 +55,25 @@ def get_best_answer(user_input):
             "🔗 https://ue.edu.pk/allfeestructure.php"
         )
 
-    # 🔁 Continue with normal similarity-based logic
+    # 🔍 Similarity matching
     user_embedding = similarity_model.encode(user_input_lower, convert_to_tensor=True)
     similarities = util.pytorch_cos_sim(user_embedding, dataset_embeddings)[0]
     best_match_idx = similarities.argmax().item()
     best_score = similarities[best_match_idx].item()
 
+    # ✏️ If not matched well, log to Excel
+    if best_score < 0.65:
+        file_path = "unmatched_queries.xlsx"
+        if os.path.exists(file_path):
+            try:
+                df = pd.read_excel(file_path)
+                new_row = {"Unmatched Queries": user_input}
+                df = pd.concat([df, pd.DataFrame([new_row])], ignore_index=True)
+                df.to_excel(file_path, index=False)
+            except Exception as e:
+                print(f"Error updating unmatched_queries.xlsx: {e}")
+
+    # 🧠 Prompt construction
     if best_score >= 0.65:
         original_answer = dataset_answers[best_match_idx]
         prompt = f"""As an official assistant for University of Education Lahore, provide a clear response:
@@ -73,8 +87,10 @@ def get_best_answer(user_input):
         Question: {user_input}
         Official Answer:"""
 
+    # 🧠 Query LLM
     llm_response = query_groq_llm(prompt)
 
+    # 🧾 Process LLM output
     if llm_response:
         for marker in ["Improved Answer:", "Official Answer:"]:
             if marker in llm_response: