config.py

import os
from dotenv import load_dotenv

load_dotenv()

class Config:
    # OpenAI Configuration
    OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
    OPENAI_MODEL = "gpt-3.5-turbo"
    EMBEDDING_MODEL = "text-embedding-ada-002"
    
    # Vector Database Configuration
    VECTOR_DB_PATH = "./vector_db"
    COLLECTION_NAME = "wueb_documents"
    
    # Chunking Configuration
    CHUNK_SIZE = 1000
    CHUNK_OVERLAP = 200
    
    # Search Configuration
    TOP_K_RESULTS = 5
    SIMILARITY_THRESHOLD = 0.7
    
    # PDF Processing
    PDF_DIRECTORY = "./pdfs"
    
    # Response Configuration
    MAX_TOKENS = 1000
    TEMPERATURE = 0.1  # Low temperature for more factual responses
    
    # System Messages
    SYSTEM_PROMPT = """You are a helpful assistant for the Wroclaw University of Economics and Business (WUEB). 
    You provide accurate information based on the university's governing documents and policies.
    
    Important guidelines:
    1. Only provide information that is explicitly stated in the provided documents
    2. If you're unsure about any information, say "I don't have enough information to answer this question accurately. Please refer to the official WUEB website or contact the university directly."
    3. Always cite the specific document or section when providing information
    4. If the question is not related to WUEB policies or documents, politely redirect to official channels
    5. Be professional and helpful in your responses"""