app.py

import gradio as gr
from langchain_community.chat_models import ChatOpenAI
from langchain.memory import ConversationBufferMemory, SimpleMemory
from langchain.agents import initialize_agent, AgentType
from langchain_community.callbacks import ClearMLCallbackHandler
from langchain_core.callbacks import StdOutCallbackHandler
from clearml import Logger, Task
from dotenv import load_dotenv
from dotenv import load_dotenv, find_dotenv
import os
import agent.planning_agent as planning_agent
import logging

# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Global variables
llm = None
chat_memory = None
query_memory = None
clearml_callback = None

def initialize_components():
    global llm, chat_memory, query_memory, clearml_callback
    load_dotenv()
    
    OPENAI_API_KEY="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"
    CLEARML_API_ACCESS_KEY="IHT6P2EMS4WMU0LGAPZ82745MXPUY3"
    CLEARML_API_SECRET_KEY="AO9UTdSwfpniravWQgfh5toQV6md7U5q6TBS70Z2Mbewrx3QWM7WxGhlyElMO3poA78"
    SERPAPI_API_KEY="619f2302253fbe56448bcf82565caf2a3263d845944682533f10b09a0d1650e6"
    
    # openai.api_key=OPENAI_API_KEY
    
    llm = ChatOpenAI(
        model_name="gpt-3.5-turbo",
        temperature=0,
        openai_api_key="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"
    )

    # Initialize memories
    chat_memory = ConversationBufferMemory(
        memory_key="chat_history",
        return_messages=True
    )
    query_memory = SimpleMemory()

    # Setup and use the ClearML Callback
    clearml_callback = ClearMLCallbackHandler(
    task_type="inference",
    project_name="langchain_callback_demo",
    task_name="llm",
    tags=["test"],
    # Change the following parameters based on the amount of detail you want tracked
    visualize=True,
    complexity_metrics=True,
    stream_logs=True,)
    
    callbacks = [StdOutCallbackHandler(), clearml_callback]
    
    # Initialize planning agent with both memories
    planning_agent.initialize_planning_agent(llm, chat_memory, query_memory, callbacks)

    logger.info("Components initialized successfully")

def process_query(query, history):
    try:
        # Restore chat history from Gradio's history
        if history:
            for human_msg, ai_msg in history:
                if chat_memory and hasattr(chat_memory, 'chat_memory'):
                    chat_memory.chat_memory.add_user_message(human_msg)
                    chat_memory.chat_memory.add_ai_message(ai_msg)
        
        # Store original query in query memory
        query_memory.memories['original_query'] = query
        
        # Execute query through planning agent
        response = planning_agent.execute(query)
        #clearml_callback.flush_tracker(name="Planning agent", finish=True)
        #clearml_callback.flush_tracker(langchain_asset=llm, name="simple_sequential")
        
        # Add current interaction to chat memory
        if chat_memory and hasattr(chat_memory, 'chat_memory'):
            chat_memory.chat_memory.add_user_message(query)
            chat_memory.chat_memory.add_ai_message(response)
        
        return response

    except Exception as e:
        error_msg = f"Error processing query: {str(e)}"
        logger.error(f"Error details: {str(e)}")

        if chat_memory and hasattr(chat_memory, 'chat_memory'):
            chat_memory.chat_memory.add_user_message(query)
            chat_memory.chat_memory.add_ai_message(error_msg)

        return error_msg

def clear_context():
    planning_agent.clear_context()
    chat_memory.clear()
    query_memory.memories.clear()
    return [], []

def create_gradio_app():
    from interface import create_interface
    return create_interface(process_query, clear_context)

def report_table(loer, iteration=0):
    # type: (Logger, int) -> ()
    """
    reporting tables to the plots section
    :param logger: The task.logger to use for sending the plots
    :param iteration: The iteration number of the current reports
    """
    # report table
    # Report table - CSV from path
    csv_path = './data/cleaned_dataset_full.csv'
    loer.report_table("Data Set Capstone", "remote csv", iteration=iteration, csv=csv_path)

def main():
    """Main application entry point"""
    try:
        initialize_components()
        app = create_gradio_app()
        app.queue()
        app.launch(server_name="0.0.0.0", server_port=7860, share=True)
        a_task = Task.get_task(project_name='langchain_callback_demo', task_name='llm')
        loer = a_task.get_logger()
        report_logs(loer)
        # report text as debug example
        report_debug_text(loer)
        # report graphs
        report_table(loer)
        loer.flush()
    except Exception as e:
        logger.error(f"Error in main: {str(e)}")
        raise

if __name__ == "__main__":
    main()