app.py

#simple chatbot con Langchain y usando Amazon Bedrock.
import gradio as gr
import random
import langchain
import langchain_community
from langchain_aws import ChatBedrock
from langchain.chains import ConversationChain
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
#para manejar la memoria del chat
import uuid
from langchain_community.chat_message_histories import ChatMessageHistory
from langchain_core.chat_history import BaseChatMessageHistory
from langchain_core.runnables.history import RunnableWithMessageHistory

#para langchain 0.2
store = {}

#funcion que retorna un id de sesion usando uuid
def get_chat_session_id():
    return str(uuid.uuid4())


def get_session_history(session_id: str) -> BaseChatMessageHistory:
    if session_id not in store:
        store[session_id] = ChatMessageHistory()
    return store[session_id]

def init():
    #definir chat con memoria

    chat = ChatBedrock(
      model_id="anthropic.claude-3-sonnet-20240229-v1:0",
      model_kwargs={"temperature": 0.1}
    )
    prompt = ChatPromptTemplate.from_messages(
      [
        (
            "system",
            "You're an assistant who's good at {ability}. Respond in 20 words or fewer",
        ),
        MessagesPlaceholder(variable_name="history"),
        ("human", "{input}"),
      ]
    )
    runnable = prompt | chat

    with_message_history = RunnableWithMessageHistory(
       runnable,
       get_session_history,
       input_messages_key="input",
       history_messages_key="history",
     )
    return with_message_history

def bedrock_response(message,history):
    the_ability= "Filosofia"
    response=the_chat.invoke(
        {"ability": the_ability, "input": message},
        config={"configurable": {"session_id": id_session}},
    )
    print(type(response))
    print(response)
    return response.content

id_session=get_chat_session_id()
the_chat = init()
demo = gr.ChatInterface(bedrock_response)

if __name__ == "__main__":
    demo.launch()