rag.py

import json
from sentence_transformers import SentenceTransformer, util
from groq import Groq
import os
import csv
from dotenv import load_dotenv

# Load environment variables
load_dotenv()

# Initialize Groq client
groq_client = Groq(api_key=os.getenv("GROQ_API_KEY"))

# Load similarity model
similarity_model = SentenceTransformer('paraphrase-MiniLM-L6-v2')

# Load dataset
with open('dataset.json', 'r', encoding='utf-8') as f:
    dataset = json.load(f)

# Precompute embeddings
dataset_questions = [item.get("input", "").lower().strip() for item in dataset]
dataset_answers = [item.get("response", "") for item in dataset]
dataset_embeddings = similarity_model.encode(dataset_questions, convert_to_tensor=True)

# Use absolute path for unmatched_queries.csv
base_dir = os.path.dirname(os.path.abspath(__file__))
file_path = os.path.join(base_dir, "unmatched_queries.csv")
print(f"[DEBUG] Writing to absolute path: {file_path}")

def query_groq_llm(prompt, model_name="llama3-70b-8192"):
    try:
        chat_completion = groq_client.chat.completions.create(
            messages=[{
                "role": "user",
                "content": prompt
            }],
            model=model_name,
            temperature=0.7,
            max_tokens=500
        )
        return chat_completion.choices[0].message.content.strip()
    except Exception as e:
        print(f"Error querying Groq API: {e}")
        return ""

def get_best_answer(user_input):
    user_input_lower = user_input.lower().strip()

    # 👉 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"
            "You’ll find comprehensive information regarding tuition, admission charges, and other applicable fees there.\n"
            "🔗 https://ue.edu.pk/allfeestructure.php"
        )

    # 🔍 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()

    # ✏️ Log to CSV if similarity is low
    if best_score < 0.65:
        print(f"[DEBUG] Similarity score too low: {best_score}. Logging query to: {file_path}")

        # Create CSV with header if it doesn't exist
        if not os.path.exists(file_path):
            print(f"[DEBUG] File {file_path} does not exist. Creating file with header.")
            try:
                with open(file_path, mode="w", newline="", encoding="utf-8") as file:
                    writer = csv.writer(file)
                    writer.writerow(["Unmatched Queries"])
                print(f"[DEBUG] Header written successfully.")
            except Exception as e:
                print(f"[ERROR] Failed to create file: {e}")

        # Append unmatched query
        try:
            with open(file_path, mode="a", newline="", encoding="utf-8") as file:
                writer = csv.writer(file)
                writer.writerow([user_input])
                print(f"[DEBUG] Query logged: {user_input}")
        except Exception as e:
            print(f"[ERROR] Failed to write query to CSV: {e}")

    # 🧠 Construct prompt
    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:
        Question: {user_input}
        Original Answer: {original_answer}
        Improved Answer:"""
    else:
        prompt = f"""As an official assistant for University of Education Lahore, provide a helpful response:
        Include relevant details about university policies.
        If unsure, direct to official channels.
        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:
                response = llm_response.split(marker)[-1].strip()
                break
        else:
            response = llm_response
    else:
        response = dataset_answers[best_match_idx] if best_score >= 0.65 else """For official information:
        📞 +92-42-99262231-33
        ✉️ [email protected]
        🌐 ue.edu.pk"""

    return response