Add chat history 添加聊天记录

背景

在许多问答应用程序中，我们希望允许用户进行来回的对话，这意味着应用程序需要对过去的问题和答案进行某种“记忆”，并将这些问题和答案纳入当前思维中。

需要做到两件事：

  • 提示：更新我们的提示，以支持历史消息作为输入。

  • 情境化问题：添加一个子链，它接受最新的用户问题，并在聊天历史的上下文中重新表达。这是需要的，以防最新的问题引用了过去消息中的某些上下文。
    例如，如果用户问一个后续问题，如“你能详细说明第二点吗？”
    ，如果没有先前消息的上下文，这是无法理解的。因此，我们不能有效地执行检索这样的问题。

实战代码

所需依赖

1
2
3
4
5
6
7
8
# BeautifulSoup (通常缩写为：BS4)  用于解析 HTML 和 XML 文件。它提供了从 HTML 和 XML 文档中提取数据的功能，以及对解析树进行导航的工具。
import bs4
from langchain import hub
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.vectorstores import Chroma
from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough
from langchain_text_splitters import RecursiveCharacterTextSplitter

初始化环境

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
import os
from langchain_openai import OpenAI
from langchain_openai import ChatOpenAI

os.environ["OPENAI_API_KEY"] = 'your key'
os.environ["OPENAI_API_BASE"] = 'https://api.moonshot.cn/v1/'

api_key = os.getenv("OPENAI_API_KEY")
base_url = os.getenv("OPENAI_API_BASE")

print("api_key=" + api_key, "base_url=" + base_url)

# 初始化模型(两种方式)：

llm = OpenAI(
    openai_api_base=base_url,
    openai_api_key=api_key,
    model_name="moonshot-v1-8k",
    temperature=0.7,
)

chat = ChatOpenAI(
    openai_api_base=base_url,
    openai_api_key=api_key,
    model_name="moonshot-v1-8k",
    temperature=0.7,
).bind(logprobs=True)

embeddings 初始化

1
2
3
from langchain.embeddings import SentenceTransformerEmbeddings

embeddings = SentenceTransformerEmbeddings(model_name="all-MiniLM-L6-v2")

Chain without chat history 无聊天记录的链

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
# 无聊天记录的链
# Load, chunk and index the contents of the blog.
loader = WebBaseLoader(
    web_paths=("https://lilianweng.github.io/posts/2023-06-23-agent/",),
    bs_kwargs=dict(
        parse_only=bs4.SoupStrainer(
            class_=("post-content", "post-title", "post-header")
        )
    ),
)
docs = loader.load()

text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
splits = text_splitter.split_documents(docs)
vectorstore = Chroma.from_documents(documents=splits, embedding=embeddings)

# Retrieve and generate using the relevant snippets of the blog.
retriever = vectorstore.as_retriever()
prompt = hub.pull("rlm/rag-prompt")


def format_docs(docs):
    return "\n\n".join(doc.page_content for doc in docs)


rag_chain = (
        {"context": retriever | format_docs, "question": RunnablePassthrough()}
        | prompt
        | chat
        | StrOutputParser()
)

rag_chain.invoke("What is Task Decomposition?")

Contextualizing the question 将问题置于背景中

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
# 将问题置于背景中
from langchain.chains import create_history_aware_retriever
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder

contextualize_q_system_prompt = """Given a chat history and the latest user question \
which might reference context in the chat history, formulate a standalone question \
which can be understood without the chat history. Do NOT answer the question, \
just reformulate it if needed and otherwise return it as is."""
contextualize_q_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", contextualize_q_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)
history_aware_retriever = create_history_aware_retriever(
    chat, retriever, contextualize_q_prompt
)

# 这个链预先对检索器的输入查询进行改写，以便检索包含会话的上下文。

Chain with chat history 带聊天记录的链

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
from langchain.chains import create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain

qa_system_prompt = """You are an assistant for question-answering tasks. \
Use the following pieces of retrieved context to answer the question. \
If you don't know the answer, just say that you don't know. \
Use three sentences maximum and keep the answer concise.\

{context}"""
qa_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", qa_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)

question_answer_chain = create_stuff_documents_chain(chat, qa_prompt)

rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain)
1
2
3
4
5
6
7
8
9
10
11
12
from langchain_core.messages import HumanMessage

chat_history = []

question = "What is Task Decomposition?"
ai_msg_1 = rag_chain.invoke({"input": question, "chat_history": chat_history})
chat_history.extend([HumanMessage(content=question), ai_msg_1["answer"]])

second_question = "What are common ways of doing it?"
ai_msg_2 = rag_chain.invoke({"input": second_question, "chat_history": chat_history})

print(ai_msg_2["answer"])

所有必要的步骤都放在一个代码单元中

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
import bs4
from langchain.chains import create_history_aware_retriever, create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain_community.chat_message_histories import ChatMessageHistory
from langchain_community.document_loaders import WebBaseLoader
from langchain_community.vectorstores import Chroma
from langchain_core.chat_history import BaseChatMessageHistory
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_text_splitters import RecursiveCharacterTextSplitter

# bs_strainer = bs4.SoupStrainer(class_=("post-content", "post-title", "post-header"))

### Construct retriever ###
loader = WebBaseLoader(
    web_paths=("https://lilianweng.github.io/posts/2023-06-23-agent/",),
    bs_kwargs=dict(
        parse_only=bs4.SoupStrainer(
            class_=("post-content", "post-title", "post-header")
        )
    ),
)
docs = loader.load()

text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
splits = text_splitter.split_documents(docs)
vectorstore = Chroma.from_documents(documents=splits, embedding=embeddings)
retriever = vectorstore.as_retriever()

### Contextualize question ###
contextualize_q_system_prompt = """Given a chat history and the latest user question \
which might reference context in the chat history, formulate a standalone question \
which can be understood without the chat history. Do NOT answer the question, \
just reformulate it if needed and otherwise return it as is."""
contextualize_q_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", contextualize_q_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)
history_aware_retriever = create_history_aware_retriever(
    chat, retriever, contextualize_q_prompt
)

### Answer question ###
qa_system_prompt = """You are an assistant for question-answering tasks. \
Use the following pieces of retrieved context to answer the question. \
If you don't know the answer, just say that you don't know. \
Use three sentences maximum and keep the answer concise.\

{context}"""
qa_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", qa_system_prompt),
        MessagesPlaceholder("chat_history"),
        ("human", "{input}"),
    ]
)
question_answer_chain = create_stuff_documents_chain(chat, qa_prompt)

rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain)

### Statefully manage chat history ###
store = {}


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


conversational_rag_chain = RunnableWithMessageHistory(
    rag_chain,
    get_session_history,
    input_messages_key="input",
    history_messages_key="chat_history",
    output_messages_key="answer",
)

conversational_rag_chain.invoke(
    {"input": "What is Task Decomposition?"},
    config={
        "configurable": {"session_id": "abc123"}
    },  # constructs a key "abc123" in `store`.
)["answer"]

conversational_rag_chain.invoke(
    {"input": "What are common ways of doing it?"},
    config={"configurable": {"session_id": "abc123"}},
)["answer"]