自然语言处理从入门到应用——LangChain:记忆(Memory)-[聊天消息记录]

分类目录:《自然语言处理从入门到应用》总目录


Cassandra聊天消息记录

Cassandra是一种分布式数据库,非常适合存储大量数据,是存储聊天消息历史的良好选择,因为它易于扩展,能够处理大量写入操作。

csharp 复制代码
# List of contact points to try connecting to Cassandra cluster.
contact_points = ["cassandra"]

from langchain.memory import CassandraChatMessageHistory

message_history = CassandraChatMessageHistory(
    contact_points=contact_points, session_id="test-session"
)

message_history.add_user_message("hi!")

message_history.add_ai_message("whats up?")
message_history.messages
[HumanMessage(content='hi!', additional_kwargs={}, example=False),
AIMessage(content='whats up?', additional_kwargs={}, example=False)]

DynamoDB聊天消息记录

首先确保我们已经正确配置了AWS CLI,并再确保我们已经安装了boto3。接下来,创建我们将存储消息 DynamoDB表:

csharp 复制代码
import boto3

# Get the service resource.
dynamodb = boto3.resource('dynamodb')

# Create the DynamoDB table.
table = dynamodb.create_table(
    TableName='SessionTable',
    KeySchema=[
        {
            'AttributeName': 'SessionId',
            'KeyType': 'HASH'
        }
    ],
    AttributeDefinitions=[
        {
            'AttributeName': 'SessionId',
            'AttributeType': 'S'
        }
    ],
    BillingMode='PAY_PER_REQUEST',
)

# Wait until the table exists.
table.meta.client.get_waiter('table_exists').wait(TableName='SessionTable')

# Print out some data about the table.
print(table.item_count)

输出:

复制代码
0
DynamoDBChatMessageHistory
csharp 复制代码
from langchain.memory.chat_message_histories import DynamoDBChatMessageHistory

history = DynamoDBChatMessageHistory(table_name="SessionTable", session_id="0")
history.add_user_message("hi!")
history.add_ai_message("whats up?")
history.messages

输出:

复制代码
[HumanMessage(content='hi!', additional_kwargs={}, example=False),
AIMessage(content='whats up?', additional_kwargs={}, example=False)]
使用自定义端点URL的DynamoDBChatMessageHistory

有时候在连接到AWS端点时指定URL非常有用,比如在本地使用Localstack进行开发。对于这种情况,我们可以通过构造函数中的endpoint_url参数来指定URL。

csharp 复制代码
from langchain.memory.chat_message_histories import DynamoDBChatMessageHistory

history = DynamoDBChatMessageHistory(table_name="SessionTable", session_id="0", endpoint_url="http://localhost.localstack.cloud:4566")
Agent with DynamoDB Memory
csharp 复制代码
from langchain.agents import Tool
from langchain.memory import ConversationBufferMemory
from langchain.chat_models import ChatOpenAI
from langchain.agents import initialize_agent
from langchain.agents import AgentType
from langchain.utilities import PythonREPL
from getpass import getpass

message_history = DynamoDBChatMessageHistory(table_name="SessionTable", session_id="1")
memory = ConversationBufferMemory(memory_key="chat_history", chat_memory=message_history, return_messages=True)
python_repl = PythonREPL()

# You can create the tool to pass to an agent
tools = [Tool(
    name="python_repl",
    description="A Python shell. Use this to execute python commands. Input should be a valid python command. If you want to see the output of a value, you should print it out with `print(...)`.",
    func=python_repl.run
)]
llm=ChatOpenAI(temperature=0)
agent_chain = initialize_agent(tools, llm, agent=AgentType.CHAT_CONVERSATIONAL_REACT_DESCRIPTION, verbose=True, memory=memory)
agent_chain.run(input="Hello!")

日志输出:

复制代码
> Entering new AgentExecutor chain...
{
    "action": "Final Answer",
    "action_input": "Hello! How can I assist you today?"
}

> Finished chain.

输出:

复制代码
'Hello! How can I assist you today?'

输入:

复制代码
agent_chain.run(input="Who owns Twitter?")

日志输出:

复制代码
> Entering new AgentExecutor chain...
{
    "action": "python_repl",
    "action_input": "import requests\nfrom bs4 import BeautifulSoup\n\nurl = 'https://en.wikipedia.org/wiki/Twitter'\nresponse = requests.get(url)\nsoup = BeautifulSoup(response.content, 'html.parser')\nowner = soup.find('th', text='Owner').find_next_sibling('td').text.strip()\nprint(owner)"
}
Observation: X Corp. (2023--present)Twitter, Inc. (2006--2023)

Thought:{
    "action": "Final Answer",
    "action_input": "X Corp. (2023--present)Twitter, Inc. (2006--2023)"
}

> Finished chain.

输出:

复制代码
'X Corp. (2023--present)Twitter, Inc. (2006--2023)'

输入:

复制代码
agent_chain.run(input="My name is Bob.")

日志输出:

复制代码
> Entering new AgentExecutor chain...
{
    "action": "Final Answer",
    "action_input": "Hello Bob! How can I assist you today?"
}

> Finished chain.

输出:

复制代码
  'Hello Bob! How can I assist you today?'

输入:

复制代码
agent_chain.run(input="Who am I?")

日志输出:

复制代码
> Entering new AgentExecutor chain...
{
    "action": "Final Answer",
    "action_input": "Your name is Bob."
}

> Finished chain.

输出:

复制代码
'Your name is Bob.'

Momento聊天消息记录

本节介绍如何使用Momento Cache来存储聊天消息记录,我们会使用MomentoChatMessageHistory类。需要注意的是,默认情况下,如果不存在具有给定名称的缓存,我们将创建一个新的缓存。我们需要获得一个Momento授权令牌才能使用这个类。这可以直接通过将其传递给momento.CacheClient实例化,作为MomentoChatMessageHistory.from_client_params的命名参数auth_token,或者可以将其设置为环境变量MOMENTO_AUTH_TOKEN

csharp 复制代码
from datetime import timedelta
from langchain.memory import MomentoChatMessageHistory

session_id = "foo"
cache_name = "langchain"
ttl = timedelta(days=1)
history = MomentoChatMessageHistory.from_client_params(
    session_id, 
    cache_name,
    ttl,
)

history.add_user_message("hi!")

history.add_ai_message("whats up?")
history.messages

输出:

复制代码
[HumanMessage(content='hi!', additional_kwargs={}, example=False),
AIMessage(content='whats up?', additional_kwargs={}, example=False)]

MongoDB聊天消息记录

本节介绍如何使用MongoDB存储聊天消息记录。MongoDB是一个开放源代码的跨平台文档导向数据库程序。它被归类为NoSQL数据库程序,使用类似JSON的文档,并且支持可选的模式。MongoDB由MongoDB Inc.开发,并在服务器端公共许可证(SSPL)下许可。

csharp 复制代码
# Provide the connection string to connect to the MongoDB database
connection_string = "mongodb://mongo_user:password123@mongo:27017"
from langchain.memory import MongoDBChatMessageHistory

message_history = MongoDBChatMessageHistory(
        connection_string=connection_string, session_id="test-session"
    )

message_history.add_user_message("hi!")

message_history.add_ai_message("whats up?")
message_history.messages

输出:

复制代码
[HumanMessage(content='hi!', additional_kwargs={}, example=False),
AIMessage(content='whats up?', additional_kwargs={}, example=False)]

Postgres聊天消息历史记录

本节介绍了如何使用 Postgres 来存储聊天消息历史记录。

csharp 复制代码
from langchain.memory import PostgresChatMessageHistory

history = PostgresChatMessageHistory(connection_string="postgresql://postgres:mypassword@localhost/chat_history", session_id="foo")

history.add_user_message("hi!")

history.add_ai_message("whats up?")
history.messages

Redis聊天消息历史记录

本节介绍了如何使用Redis来存储聊天消息历史记录。

csharp 复制代码
from langchain.memory import RedisChatMessageHistory

history = RedisChatMessageHistory("foo")

history.add_user_message("hi!")
history.add_ai_message("whats up?")
history.messages

输出:

复制代码
[AIMessage(content='whats up?', additional_kwargs={}),
HumanMessage(content='hi!', additional_kwargs={})]

参考文献:

1 LangChain官方网站:https://www.langchain.com/

2 LangChain 🦜️🔗 中文网,跟着LangChain一起学LLM/GPT开发:https://www.langchain.com.cn/

3 LangChain中文网 - LangChain 是一个用于开发由语言模型驱动的应用程序的框架:http://www.cnlangchain.com/

相关推荐
蝎子莱莱爱打怪15 分钟前
Claude Code 官宣新升级:子智能体默认后台跑,你边聊它边干活
人工智能
武子康20 分钟前
调查研究-206 DeepSeek DSpark 深度解析:大模型推理加速,正在从“模型能力”转向“系统工程”
人工智能·agent·deepseek
花千树_01027 分钟前
多工具调用只是开始:用 Regnexe 构建真正会反思的 Java Agent
langchain·agent
甲维斯1 小时前
最佳work模型sonnet5来了,直接就能用!
人工智能
IT_陈寒1 小时前
React hooks 闭包陷阱把我的状态吃掉了,原来问题出在这里
前端·人工智能·后端
冬奇Lab13 小时前
Workflow 系列(03):状态管理——持久化、幂等性与版本绑定
人工智能·工作流引擎
冬奇Lab13 小时前
每日一个开源项目(第146篇):openpilot - 开源自动驾驶辅助系统,曾在 Consumer Reports 评测中超过特斯拉 Autopilot
人工智能·开源·自动驾驶
吴佳浩15 小时前
AI 工程师知识地图:模型格式、框架、部署工具一次讲明白
人工智能·aigc·ai编程