# Import necessary libraries import os # Interacting with the operating system (reading/writing files) import chromadb # High-performance vector database for storing/querying dense vectors from dotenv import load_dotenv # Loading environment variables from a .env file import json # Parsing and handling JSON data # LangChain imports from langchain_openai import ChatOpenAI from langchain_core.documents import Document # Document data structures from langchain_core.runnables import RunnablePassthrough # LangChain core library for running pipelines from langchain_core.output_parsers import StrOutputParser # String output parser from langchain.prompts import ChatPromptTemplate # Template for chat prompts from langchain.chains.query_constructor.base import AttributeInfo # Base classes for query construction from langchain.retrievers.self_query.base import SelfQueryRetriever # Base classes for self-querying retrievers from langchain.retrievers.document_compressors import LLMChainExtractor, CrossEncoderReranker # Document compressors from langchain.retrievers import ContextualCompressionRetriever # Contextual compression retrievers # LangChain community & experimental imports from langchain_community.vectorstores import Chroma # Implementations of vector stores like Chroma from langchain_community.document_loaders import PyPDFDirectoryLoader, PyPDFLoader # Document loaders for PDFs from langchain_community.cross_encoders import HuggingFaceCrossEncoder # Cross-encoders from HuggingFace from langchain_experimental.text_splitter import SemanticChunker # Experimental text splitting methods from langchain.text_splitter import ( CharacterTextSplitter, # Splitting text by characters RecursiveCharacterTextSplitter # Recursive splitting of text by characters ) from langchain_core.tools import tool from langchain.agents import create_tool_calling_agent, AgentExecutor from langchain_core.prompts import ChatPromptTemplate # LangChain OpenAI imports from langchain_openai import AzureOpenAIEmbeddings, AzureChatOpenAI # OpenAI embeddings and models from langchain.embeddings.openai import OpenAIEmbeddings # OpenAI embeddings for text vectors # LlamaParse & LlamaIndex imports from llama_parse import LlamaParse # Document parsing library from llama_index.core import Settings, SimpleDirectoryReader # Core functionalities of the LlamaIndex # LangGraph import from langgraph.graph import StateGraph, END, START # State graph for managing states in LangChain # Pydantic import from pydantic import BaseModel # Pydantic for data validation # Typing imports from typing import Dict, List, Tuple, Any, TypedDict # Python typing for function annotations # Other utilities import numpy as np # Numpy for numerical operations from groq import Groq from mem0 import MemoryClient import streamlit as st from datetime import datetime #====================================SETUP=====================================# # Fetch secrets from Hugging Face Spaces api_key = os.environ.get("API_KEY") or config.get("API_KEY") endpoint = os.environ.get("OPENAI_API_BASE") or config.get("OPENAI_API_BASE") llama_api_key = os.environ.get("GROQ_API_KEY") or config2.get("LLAMA_KEY") MEM0_api_key = os.environ.get("MEM0_API_KEY") # Initialize the OpenAI Embeddings embedding_model = OpenAIEmbeddings( openai_api_base=endpoint, openai_api_key=api_key, model='text-embedding-ada-002' ) # Initialize the Chat OpenAI model llm = ChatOpenAI( openai_api_base=endpoint, openai_api_key=api_key, model="gpt-4o-mini", streaming=False ) # This initializes the Chat OpenAI model with the provided endpoint, API key, deployment name, and a temperature setting of 0 (to control response variability). # set the LLM and embedding model in the LlamaIndex settings. Settings.llm = llm Settings.embedding = embedding_model #================================Creating Langgraph agent======================# class AgentState(TypedDict): query: str # The current user query expanded_query: str # The expanded version of the user query context: List[Dict[str, Any]] # Retrieved documents (content and metadata) response: str # The generated response to the user query precision_score: float # The precision score of the response groundedness_score: float # The groundedness score of the response groundedness_loop_count: int # Counter for groundedness refinement loops precision_loop_count: int # Counter for precision refinement loops feedback: str query_feedback: str groundedness_check: bool loop_max_iter: int def expand_query(state: Dict[str, Any]) -> Dict[str, Any]: """ Expands the user query to improve retrieval of nutrition-disorder information. Args: state: Workflow state containing at least 'query' and 'query_feedback'. Returns: Workflow state with an additional 'expanded_query' key. """ s: AgentState = state print("---------Expanding Query---------") system_message = '''You are an assistant that reformulates vague or short user questions into detailed, domain-specific queries related to nutrition disorders. Examples: - Input: "What about iron?" Expanded: "What are the common symptoms and treatments for iron deficiency anemia?" - Input: "Diets for gut issues?" Expanded: "What dietary recommendations are effective for managing irritable bowel syndrome and promoting gut microbiome health?" - Input: "Sugar" Expanded: "What are the risks of high sugar intake in diabetic patients and how can it be managed nutritionally?" ''' expand_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Expand this query: {query} using the feedback: {query_feedback}") ]) chain = expand_prompt | llm | StrOutputParser() expanded_query = chain.invoke({"query": state['query'], "query_feedback":state["query_feedback"]}) print("expanded_query", expanded_query) state["expanded_query"] = expanded_query return state chroma_client = chromadb.PersistentClient(path="./combined") vector_store = Chroma( client=chroma_client, # <- pass the client you just made collection_name="combined", embedding_function=embedding_model, ) # Create a retriever from the vector store retriever = vector_store.as_retriever( search_type='similarity', search_kwargs={'k': 3} ) def retrieve_context(state): """ Retrieves context from the vector store using the expanded or original query. Args: state (Dict): The current state of the workflow, containing the query and expanded query. Returns: Dict: The updated state with the retrieved context. """ print("---------retrieve_context---------") query = state['expanded_query'] # Complete the code to define the key for the expanded query #print("Query used for retrieval:", query) # Debugging: Print the query # Retrieve documents from the vector store docs = retriever.invoke(query) print("Retrieved documents:", docs) # Debugging: Print the raw docs object # Extract both page_content and metadata from each document context = [ { "content": doc.metadata.get("original_content", doc.page_content), "metadata": doc.metadata } for doc in docs ] state['context'] = context # Complete the code to define the key for storing the context print("Extracted context with metadata:", context) # Debugging: Print the extracted context #print(f"Groundedness loop count: {state['groundedness_loop_count']}") return state def craft_response(state): """ Generates a response using the retrieved context, focusing on nutrition disorders. Args: state (Dict): The current state of the workflow, containing the query and retrieved context. Returns: Dict: The updated state with the generated response. """ print("---------craft_response---------") system_message = '''You are a helpful AI assistant trained to support healthcare providers in retrieving accurate and relevant information related to nutrition disorders. Your responses must strictly adhere to the retrieved context, which is extracted from the Nutritional Medical Reference or similar trusted sources. Do not speculate or introduce external knowledge. Focus only on symptoms, diagnoses, treatment plans, or other clinical details found in the context. If the context does not contain enough information to answer accurately, clearly state that. Aim for clarity, factual grounding, and relevance to the user's query.''' response_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Query: {query}\nContext: {context}\n\nfeedback: {feedback}") ]) chain = response_prompt | llm response = chain.invoke({ "query": state['query'], "context": "\n".join([doc["content"] for doc in state['context']]), "feedback": state.get('feedback', 'No feedback provided') # add feedback to the prompt }) state['response'] = response print("intermediate response: ", response) return state def score_groundedness(state): """ Checks whether the response is grounded in the retrieved context. Args: state (Dict): The current state of the workflow, containing the response and context. Returns: Dict: The updated state with the groundedness score. """ print("---------check_groundedness---------") system_message = '''You are evaluating whether an AI-generated response is grounded in the retrieved context provided from nutritional health documents. The context includes evidence and facts, and your task is to assign a groundedness score between 0 and 1, where: - 1.0 means the response is fully supported by the context, - 0.0 means the response is entirely unsupported. Be strict: if even a part of the response is not traceable to the context, reduce the score. Provide only the numeric score.''' groundedness_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Context: {context}\nResponse: {response}\n\nGroundedness score:") ]) chain = groundedness_prompt | llm | StrOutputParser() groundedness_score = float(chain.invoke({ "context": "\n".join([doc["content"] for doc in state['context']]), "response":state['response'] # Complete the code to define the response })) print("groundedness_score: ", groundedness_score) state['groundedness_loop_count'] += 1 print("#########Groundedness Incremented###########") state['groundedness_score'] = groundedness_score return state def check_precision(state: Dict) -> Dict: """ Checks whether the response precisely addresses the user’s query. Args: state (Dict): The current state of the workflow, containing the query and response. Returns: Dict: The updated state with the precision score. """ print("---------check_precision---------") system_message = '''You are assessing whether an AI-generated response precisely answers the user's query, especially within the domain of nutritional health and disorders. Provide a precision score between 0 and 1: - 1.0: The response fully and directly answers the query with clear relevance. - 0.0: The response is vague, unrelated, or fails to address the query. Only return a numeric score.''' precision_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Query: {query}\nResponse: {response}\n\nPrecision score:") ]) chain = precision_prompt | llm | StrOutputParser() # Complete the code to define the chain of processing precision_score = float(chain.invoke({ "query": state['query'], "response":state['response'] # Complete the code to access the response from the state })) state['precision_score'] = precision_score print("precision_score:", precision_score) state['precision_loop_count'] +=1 print("#########Precision Incremented###########") return state def refine_response(state: Dict) -> Dict: """ Suggests improvements for the generated response. Args: state (Dict): The current state of the workflow, containing the query and response. Returns: Dict: The updated state with response refinement suggestions. """ print("---------refine_response---------") system_message = '''You are an expert assistant helping to improve AI-generated answers related to nutritional disorders. Evaluate the response and suggest constructive improvements to enhance accuracy, specificity, and completeness. Do not rewrite the response. Instead, point out what is vague, missing, or could be better explained. Focus on clinical terminology, nutritional details, and relevance to the user query.''' refine_response_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Query: {query}\nResponse: {response}\n\n" "What improvements can be made to enhance accuracy and completeness?") ]) chain = refine_response_prompt | llm| StrOutputParser() # Store response suggestions in a structured format feedback = f"Previous Response: {state['response']}\nSuggestions: {chain.invoke({'query': state['query'], 'response': state['response']})}" print("feedback: ", feedback) print(f"State: {state}") state['feedback'] = feedback return state def refine_query(state: Dict) -> Dict: """ Suggests improvements for the expanded query. Args: state (Dict): The current state of the workflow, containing the query and expanded query. Returns: Dict: The updated state with query refinement suggestions. """ print("---------refine_query---------") system_message = '''You are an expert retrieval assistant helping to improve search queries related to nutrition disorders. Analyze the original and expanded queries and provide suggestions to increase search precision. Focus on identifying missing clinical terms, relevant nutritional keywords, or clarifying the scope of the query. Do not rewrite the query. Provide suggestions only.''' refine_query_prompt = ChatPromptTemplate.from_messages([ ("system", system_message), ("user", "Original Query: {query}\nExpanded Query: {expanded_query}\n\n" "What improvements can be made for a better search?") ]) chain = refine_query_prompt | llm | StrOutputParser() # Store refinement suggestions without modifying the original expanded query query_feedback = f"Previous Expanded Query: {state['expanded_query']}\nSuggestions: {chain.invoke({'query': state['query'], 'expanded_query': state['expanded_query']})}" print("query_feedback: ", query_feedback) print(f"Groundedness loop count: {state['groundedness_loop_count']}") state['query_feedback'] = query_feedback return state def should_continue_groundedness(state): """Decides if groundedness is sufficient or needs improvement.""" print("---------should_continue_groundedness---------") print("groundedness loop count: ", state['groundedness_loop_count']) # Threshold logic: groundedness score should be at least 0.8 if state['groundedness_score'] >= 0.8: print("Moving to precision") return "check_precision" else: # Allow a maximum of 2 refinement loops if state['groundedness_loop_count'] > state['loop_max_iter']: print("Maximum groundedness iterations reached") return "max_iterations_reached" else: print("---------Groundedness Score Threshold Not Met. Refining Response-----------") return "refine_response" def should_continue_precision(state: Dict) -> str: """Decides if precision is sufficient or needs improvement.""" print("---------should_continue_precision---------") print("precision loop count: ", state['precision_loop_count']) # Threshold for acceptable precision score if state['precision_score'] >= 0.8: return "pass" # Complete the workflow else: # Check if maximum refinement attempts have been reached if state['precision_loop_count'] > state['loop_max_iter']: return "max_iterations_reached" else: print("---------Precision Score Threshold Not met. Refining Query-----------") return "refine_query" def max_iterations_reached(state: Dict) -> Dict: """Handles the case where max iterations are reached.""" print("---------max_iterations_reached---------") state['response'] = "We need more context to provide an accurate answer." return state from langgraph.graph import END, StateGraph, START def create_workflow() -> StateGraph: """Creates the updated workflow for the AI nutrition agent.""" workflow = StateGraph(dict) # Workflow state is a dictionary # Add processing nodes workflow.add_node("expand_query", expand_query) # Step 1: Expand user query. Complete with the function to expand the query workflow.add_node("retrieve_context", retrieve_context) # Step 2: Retrieve relevant documents. Complete with the function to retrieve context workflow.add_node("craft_response", craft_response) # Step 3: Generate a response based on retrieved data. Complete with the function to craft a response workflow.add_node("score_groundedness", score_groundedness) # Step 4: Evaluate response grounding. Complete with the function to score groundedness workflow.add_node("refine_response", refine_response) # Step 5: Improve response if it's weakly grounded. Complete with the function to refine the response workflow.add_node("check_precision", check_precision) # Step 6: Evaluate response precision. Complete with the function to check precision workflow.add_node("refine_query", refine_query) # Step 7: Improve query if response lacks precision. Complete with the function to refine the query workflow.add_node("max_iterations_reached", max_iterations_reached) # Step 8: Handle max iterations. Complete with the function to handle max iterations # Main flow edges workflow.add_edge(START, "expand_query") workflow.add_edge("expand_query", "retrieve_context") workflow.add_edge("retrieve_context", "craft_response") workflow.add_edge("craft_response", "score_groundedness") # Groundedness logic workflow.add_conditional_edges( "score_groundedness", should_continue_groundedness, { "check_precision": "check_precision", "refine_response": "refine_response", "max_iterations_reached": "max_iterations_reached" } ) # Edge to reprocess refined response workflow.add_edge("refine_response", "craft_response") # Precision logic workflow.add_conditional_edges( "check_precision", should_continue_precision, { "pass": END, "refine_query": "refine_query", "max_iterations_reached": "max_iterations_reached" } ) # Edge to re-expand refined query and reenter flow workflow.add_edge("refine_query", "expand_query") workflow.add_edge("max_iterations_reached", END) return workflow #=========================== Defining the agentic rag tool ====================# WORKFLOW_APP = create_workflow().compile() @tool def agentic_rag(query: str) -> Dict[str, Any]: """ Runs the RAG-based agent with conversation history for context-aware responses. """ if not query or not isinstance(query, str): return {"error": "Invalid or empty query provided"} inputs = { "query": query, "expanded_query": "", #Initialized as an empty string since the expand_query function will populate this field with the reformulated query based on the original query "context": [], # Retrieved documents (initially empty) "response": "", #Initialized as an empty string since the craft_response function will generate the AI response and store it here "precision_score": 0.0, #Initialized as 0.0 since the check_precision function will compute and assign a precision score between 0 and 1. "groundedness_score": 0.0, #Initialized as 0.0 since the score_groundedness function will compute and assign a groundedness score between 0 and 1. "groundedness_loop_count": 0, #Initialized as 0 to track the number of groundedness refinement loops, incremented in score_groundedness. "precision_loop_count": 0, #Initialized as 0 to track the number of precision refinement loops, incremented in check_precision. "feedback": "", #Initialized as an empty string since the refine_response function will populate this with suggestions for improving the response. "query_feedback": "", #Initialized as an empty string since the refine_query function will populate this with suggestions for improving the expanded query. "loop_max_iter": 5 } output = WORKFLOW_APP.invoke(inputs) return output #================================ Guardrails ===========================# llama_guard_client = Groq(api_key=llama_api_key) # Function to filter user input with Llama Guard def filter_input_with_llama_guard(user_input, model="meta-llama/llama-guard-4-12b"): """ Filters user input using Llama Guard to ensure it is safe. Parameters: - user_input: The input provided by the user. - model: The Llama Guard model to be used for filtering (default is "meta-llama/llama-guard-4-12bb"). Returns: - The filtered and safe input. """ try: # Create a request to Llama Guard to filter the user input response = llama_guard_client.chat.completions.create( messages=[{"role": "user", "content": user_input}], model=model, ) # Return the filtered input return response.choices[0].message.content.strip() except Exception as e: print(f"Error with Llama Guard: {e}") return None #============================= Adding Memory to the agent using mem0 ===============================# class NutritionBot: def __init__(self): """ Initialize the NutritionBot class, setting up memory, the LLM client, tools, and the agent executor. """ # Initialize a memory client to store and retrieve customer interactions self.memory = MemoryClient(api_key=os.environ.get("MEM0_API_KEY")) # Complete the code to define the memory client API key # Initialize the OpenAI client using the provided credentials self.client = ChatOpenAI( model_name="gpt-4o-mini", # Specify the model to use (e.g., GPT-4 optimized version) api_key=os.environ.get("API_KEY"), # API key for authentication openai_api_base = os.environ.get("OPENAI_API_BASE"), temperature=0 # Controls randomness in responses; 0 ensures deterministic results ) # Define tools available to the chatbot, such as web search tools = [agentic_rag] # Define the system prompt to set the behavior of the chatbot system_prompt = """You are a caring and knowledgeable Medical Support Agent, specializing in nutrition disorder-related guidance. Your goal is to provide accurate, empathetic, and tailored nutritional recommendations while ensuring a seamless customer experience. Guidelines for Interaction: Maintain a polite, professional, and reassuring tone. Show genuine empathy for customer concerns and health challenges. Reference past interactions to provide personalized and consistent advice. Engage with the customer by asking about their food preferences, dietary restrictions, and lifestyle before offering recommendations. Ensure consistent and accurate information across conversations. If any detail is unclear or missing, proactively ask for clarification. Always use the agentic_rag tool to retrieve up-to-date and evidence-based nutrition insights. Keep track of ongoing issues and follow-ups to ensure continuity in support. Your primary goal is to help customers make informed nutrition decisions that align with their health conditions and personal preferences. """ # Build the prompt template for the agent prompt = ChatPromptTemplate.from_messages([ ("system", system_prompt), # System instructions ("human", "{input}"), # Placeholder for human input ("placeholder", "{agent_scratchpad}") # Placeholder for intermediate reasoning steps ]) # Create an agent capable of interacting with tools and executing tasks agent = create_tool_calling_agent(self.client, tools, prompt) # Wrap the agent in an executor to manage tool interactions and execution flow self.agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True) def store_customer_interaction(self, user_id: str, message: str, response: str, metadata: Dict = None): """ Store customer interaction in memory for future reference. Args: user_id (str): Unique identifier for the customer. message (str): Customer's query or message. response (str): Chatbot's response. metadata (Dict, optional): Additional metadata for the interaction. """ if metadata is None: metadata = {} # Add a timestamp to the metadata for tracking purposes metadata["timestamp"] = datetime.now().isoformat() # Format the conversation for storage conversation = [ {"role": "user", "content": message}, {"role": "assistant", "content": response} ] # Store the interaction in the memory client self.memory.add( conversation, user_id=user_id, output_format="v1.1", metadata=metadata ) def get_relevant_history(self, user_id: str, query: str) -> List[Dict]: """ Retrieve past interactions relevant to the current query. Args: user_id (str): Unique identifier for the customer. query (str): The customer's current query. Returns: List[Dict]: A list of relevant past interactions. """ return self.memory.search( query=query, # Search for interactions related to the query user_id=user_id, # Restrict search to the specific user limit=5 # Complete the code to define the limit for retrieved interactions ) def handle_customer_query(self, user_id: str, query: str) -> str: """ Process a customer's query and provide a response, taking into account past interactions. Args: user_id (str): Unique identifier for the customer. query (str): Customer's query. Returns: str: Chatbot's response. """ # Retrieve relevant past interactions for context relevant_history = self.get_relevant_history(user_id, query) # Build a context string from the relevant history context = "Previous relevant interactions:\n" for memory in relevant_history: context += f"Customer: {memory['memory']}\n" # Customer's past messages context += f"Support: {memory['memory']}\n" # Chatbot's past responses context += "---\n" # Print context for debugging purposes print("Context: ", context) # Prepare a prompt combining past context and the current query prompt = f""" Context: {context} Current customer query: {query} Provide a helpful response that takes into account any relevant past interactions. """ # Generate a response using the agent response = self.agent_executor.invoke({"input": prompt}) # Store the current interaction for future reference self.store_customer_interaction( user_id=user_id, message=query, response=response["output"], metadata={"type": "support_query"} ) # Return the chatbot's response return response['output'] #=====================User Interface using streamlit ===========================# def nutrition_disorder_streamlit(): """ A Streamlit-based UI for the Nutrition Disorder Specialist Agent. """ st.title("Nutrition Disorder Specialist") st.write("Ask me anything about nutrition disorders, symptoms, causes, treatments, and more.") st.write("Type 'exit' to end the conversation.") # Initialize session state for chat history and user_id if they don't exist if 'chat_history' not in st.session_state: st.session_state.chat_history = [] if 'user_id' not in st.session_state: st.session_state.user_id = None # Login form: Only if user is not logged in if st.session_state.user_id is None: with st.form("login_form", clear_on_submit=True): user_id = st.text_input("Please enter your name to begin:") submit_button = st.form_submit_button("Login") if submit_button and user_id: st.session_state.user_id = user_id st.session_state.chat_history.append({ "role": "assistant", "content": f"Welcome, {user_id}! How can I help you with nutrition disorders today?" }) st.session_state.login_submitted = True if st.session_state.get("login_submitted", False): st.session_state.pop("login_submitted") st.rerun() else: for message in st.session_state.chat_history: with st.chat_message(message["role"]): st.write(message["content"]) # === Filled Blanks === user_query = st.chat_input("Type your question here (or 'exit' to end)...") if user_query: if user_query.lower() == "exit": st.session_state.chat_history.append({"role": "user", "content": "exit"}) with st.chat_message("user"): st.write("exit") goodbye_msg = "Goodbye! Feel free to return if you have more questions about nutrition disorders." st.session_state.chat_history.append({"role": "assistant", "content": goodbye_msg}) with st.chat_message("assistant"): st.write(goodbye_msg) st.session_state.user_id = None st.rerun() return st.session_state.chat_history.append({"role": "user", "content": user_query}) with st.chat_message("user"): st.write(user_query) filtered_result = filter_input_with_llama_guard(user_query) filtered_result = filtered_result.replace("\n", " ") if filtered_result in ["safe", "unsafe S7", "unsafe S6"]: try: if 'chatbot' not in st.session_state: st.session_state.chatbot = NutritionBot() response = st.session_state.chatbot.handle_customer_query( st.session_state.user_id, user_query) st.write(response) st.session_state.chat_history.append({"role": "assistant", "content": response}) except Exception as e: error_msg = f"Sorry, I encountered an error while processing your query. Please try again. Error: {str(e)}" st.write(error_msg) st.session_state.chat_history.append({"role": "assistant", "content": error_msg}) else: inappropriate_msg = "I apologize, but I cannot process that input as it may be inappropriate. Please try again." st.write(inappropriate_msg) st.session_state.chat_history.append({"role": "assistant", "content": inappropriate_msg}) if __name__ == "__main__": nutrition_disorder_streamlit()