Update rag.py

rag.py

@@ -1,13 +1,11 @@
 import json
 from sentence_transformers import SentenceTransformer, util
 from groq import Groq
-import datetime
-import requests
-from io import BytesIO
-from PIL import Image, ImageDraw, ImageFont
-import numpy as np
-from dotenv import load_dotenv
+from datetime import datetime
 import os
+import pandas as pd
+from datasets import load_dataset, Dataset
+from dotenv import load_dotenv
 
 # Load environment variables
 load_dotenv()
@@ -15,18 +13,53 @@ load_dotenv()
 # Initialize Groq client
 groq_client = Groq(api_key=os.getenv("GROQ_API_KEY"))
 
-# Load models and dataset
+# Load similarity model
 similarity_model = SentenceTransformer('paraphrase-MiniLM-L6-v2')
 
-# Load dataset (automatically using the path)
-with open('dataset.json', 'r') as f:
-    dataset = json.load(f)
+# Config
+HF_DATASET_REPO = "midrees2806/unmatched_queries"
+HF_TOKEN = os.getenv("HF_TOKEN")
+
+# Greeting list
+GREETINGS = [
+    "hi", "hello", "hey", "good morning", "good afternoon", "good evening",
+    "assalam o alaikum", "salam", "aoa", "hi there",
+    "hey there", "greetings"
+]
+
+# Load local dataset
+try:
+    with open('dataset.json', 'r') as f:
+        dataset = json.load(f)
+    if not all(isinstance(item, dict) and 'Question' in item and 'Answer' in item for item in dataset):
+        raise ValueError("Invalid dataset structure")
+except Exception as e:
+    print(f"Error loading dataset: {e}")
+    dataset = []
 
 # Precompute embeddings
-dataset_questions = [item.get("input", "").lower().strip() for item in dataset]
-dataset_answers = [item.get("response", "") for item in dataset]
+dataset_questions = [item.get("Question", "").lower().strip() for item in dataset]
+dataset_answers = [item.get("Answer", "") for item in dataset]
 dataset_embeddings = similarity_model.encode(dataset_questions, convert_to_tensor=True)
 
+# Save unmatched queries to Hugging Face
+def manage_unmatched_queries(query: str):
+    try:
+        timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        try:
+            ds = load_dataset(HF_DATASET_REPO, token=HF_TOKEN)
+            df = ds["train"].to_pandas()
+        except:
+            df = pd.DataFrame(columns=["Query", "Timestamp", "Processed"])
+        if query not in df["Query"].values:
+            new_entry = {"Query": query, "Timestamp": timestamp, "Processed": False}
+            df = pd.concat([df, pd.DataFrame([new_entry])], ignore_index=True)
+            updated_ds = Dataset.from_pandas(df)
+            updated_ds.push_to_hub(HF_DATASET_REPO, token=HF_TOKEN)
+    except Exception as e:
+        print(f"Failed to save query: {e}")
+
+# Query Groq LLM
 def query_groq_llm(prompt, model_name="llama3-70b-8192"):
     try:
         chat_completion = groq_client.chat.completions.create(
@@ -43,23 +76,38 @@ 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
 def get_best_answer(user_input):
+    if not user_input.strip():
+        return "Please enter a valid question."
+
     user_input_lower = user_input.lower().strip()
 
-    # 👉 Check if question is about fee
-    if any(keyword in user_input_lower for keyword in ["semester fee","semester fees"]):
+    if len(user_input_lower.split()) < 3 and not any(greet in user_input_lower for greet in GREETINGS):
+        return "Please ask your question properly with at least 3 words."
+
+    if any(greet in user_input_lower for greet in GREETINGS):
+        greeting_response = query_groq_llm(
+            f"You are an official assistant for University of Education Lahore. "
+            f"Respond to this greeting in a friendly and professional manner: {user_input}"
+        )
+        return greeting_response if greeting_response else "Hello! How can I assist you today?"
+
+    if any(keyword in user_input_lower for keyword in ["fee structure", "fees structure", "semester fees", "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"
-            "You’ll find comprehensive information regarding tuition, admission charges, and other applicable fees there.\n"
+            "You'll find comprehensive information regarding tuition, admission charges, and other applicable fees there.\n"
             "🔗 https://ue.edu.pk/allfeestructure.php"
         )
 
-    # 🔁 Continue with normal similarity-based logic
     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 best_score < 0.65:
+        manage_unmatched_queries(user_input)
+
     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:
@@ -76,16 +124,14 @@ def get_best_answer(user_input):
     llm_response = query_groq_llm(prompt)
 
     if llm_response:
-        for marker in ["Improved Answer:", "Official Answer:"]:
+        for marker in ["Improved Answer:", "Official Answer:", "Rephrased Answer:"]:
             if marker in llm_response:
-                response = llm_response.split(marker)[-1].strip()
-                break
-        else:
-            response = llm_response
+                return llm_response.split(marker)[-1].strip()
+        return llm_response
     else:
-        response = dataset_answers[best_match_idx] if best_score >= 0.65 else """For official information:
-        📞 +92-42-99262231-33
-        ✉️ info@ue.edu.pk
-        🌐 ue.edu.pk"""
-
-    return response
+        return dataset_answers[best_match_idx] if best_score >= 0.65 else (
+            "For official information:\n"
+            "📞 +92-42-99262231-33\n"
+            "✉️ info@ue.edu.pk\n"
+            "🌐 https://ue.edu.pk"
+        )