Chat Messages | 🦜️🔗 Langchain

ChatMessageHistory：基类，保存HumanMessages、AIMessages

from langchain.memory import ChatMessageHistory
history = ChatMessageHistory()
history.add_user_message("hi!")
history.add_ai_message("whats up?")
print(history.messages)
#[HumanMessage(content='hi!'), AIMessage(content='whats up?')]

Conversation Buffer | 🦜️🔗 Langchain

ConversationBufferMemory，默认return_messages = False。存储为字符串，True存储为message。详细参考langchain学习笔记（十一）-CSDN博客-（3）实践使用-ConversationBufferMemory

from langchain.memory import ConversationBufferMemory
# 默认return_messages = False
memory=ConversationBufferMemory()
memory.save_context({"input":'hi'},{"output":"what's up"})
memory.save_context({"input":'hi'},{"output":"what's up"})
# 存储为字符串
print(memory.load_memory_variables({}))
memory = ConversationBufferMemory(return_messages=True)
memory.save_context({"input": 'hi'}, {"output": "what's up"})
memory.save_context({"input": 'hi'}, {"output": "what's up"})
# 存储为message
print(memory.load_memory_variables({}))
####chain###
from langchain_openai import OpenAI
from langchain.chains import ConversationChain
llm=OpenAI(temperature=0)
conversation=ConversationChain(
    llm=llm,verbose=True,memory=ConversationBufferMemory()
)
res1=conversation.predict(input='王阳明是哪个朝代的人')
print(res1)
res2=conversation.predict(input="同个朝代还有哪些名人，简要介绍其中三个")
print(res2)

Conversation Buffer Window | 🦜️🔗 Langchain

ConversationBufferWindowMemory仅保留最近K个对话，使得缓存不会太大。

相较ConversationBufferMemory仅需要多传入一个参数K(默认k=5)，控制需要保留的对话数。其他一致

---------------------------------------------

待续