Streamlit 应用程序中带有 ConversationBufferMemory 的 LangChain 不起作用

Question

我有一个流式聊天机器人，它工作得很好，但不记得以前的聊天历史记录。我试图用 langchain ConversationBufferMemory 添加它，但它似乎不起作用。

这是我创建的聊天机器人的示例：

import streamlit as st
from streamlit_chat import message

from langchain.chains import ConversationChain
from langchain.llms import OpenAI
from langchain.chat_models import AzureChatOpenAI

from langchain.memory import ConversationBufferMemory

from langchain.prompts import (
    ChatPromptTemplate, 
    MessagesPlaceholder, 
    SystemMessagePromptTemplate, 
    HumanMessagePromptTemplate
)


prompt = ChatPromptTemplate.from_messages([
    SystemMessagePromptTemplate.from_template("The following is a friendly conversation between a human and an AI. The AI is talkative and provides lots of specific details from its context. If the AI does not know the answer to a question, it truthfully says it does not know."),
    MessagesPlaceholder(variable_name="history"),
    HumanMessagePromptTemplate.from_template("{input}")
])


def load_chain(prompt):
    """Logic for loading the chain you want to use should go here."""
    llm = AzureChatOpenAI(
                    deployment_name = 'gpt-35-turbo',
                    model_name = 'gpt-35-turbo',
                    temperature = 0,
                    openai_api_key = '.....',
                    openai_api_base = '.....',
                    openai_api_version = "2023-05-15", 
                    openai_api_type="azure"
                    )
    memory = ConversationBufferMemory(return_messages=True)
    chain = ConversationChain(
        llm=llm,
        verbose=True,
        prompt=prompt,
        memory=memory
    )
    return chain

chain = load_chain(prompt)

# From here down is all the StreamLit UI.
st.set_page_config(page_title="LangChain Demo", page_icon=":robot:")
st.header("LangChain Demo")

if "generated" not in st.session_state:
    st.session_state["generated"] = []

if "past" not in st.session_state:
    st.session_state["past"] = []

if "history" not in st.session_state:
    st.session_state["history"] = []

def get_text():
    input_text = st.text_input("You: ", "Hello, how are you?", key="input")
    return input_text


user_input = get_text()

if user_input:
    output = chain.run(input=user_input, history=st.session_state["history"])
    st.session_state["history"].append((user_input, output))
    st.session_state.past.append(user_input)
    st.session_state.generated.append(output)
    st.write(st.session_state["history"])

if st.session_state["generated"]:

    for i in range(len(st.session_state["generated"]) - 1, -1, -1):
        message(st.session_state["generated"][i], key=str(i))
        message(st.session_state["past"][i], is_user=True, key=str(i) + "_user")

看起来机器人出于某种原因忽略了 ConversationBufferMemory。任何帮助，将不胜感激。

Answer 1

@hiper2d 的回答已经说得很清楚了。在过去的一个小时里我面临着同样的问题。事后看来，我应该知道。

每当屏幕上需要更新某些内容时，Streamlit 都会从上到下重新运行整个 Python 脚本。这意味着，ConversationalBufferMemory每次刷新都会启动。如果我们要将其存储在会话状态中，我看不到内存抽象的使用。

但是，如果您确实想这样做，您可以设置会话状态：

if "messages" not in st.session_state:
    st.session_state.messages = []

# While processing
st.session_state.messages.append({"role" : "assistant", "content" : output_from_llm})