langchain/templates/rag-conversation/rag_conversation/chain.py

126 lines
4.2 KiB
Python

import os
from operator import itemgetter
from typing import List, Tuple
from langchain_community.chat_models import ChatOpenAI
from langchain_community.embeddings import OpenAIEmbeddings
from langchain_core.messages import AIMessage, HumanMessage
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import (
ChatPromptTemplate,
MessagesPlaceholder,
format_document,
)
from langchain_core.prompts.prompt import PromptTemplate
from langchain_core.pydantic_v1 import BaseModel, Field
from langchain_core.runnables import (
RunnableBranch,
RunnableLambda,
RunnableParallel,
RunnablePassthrough,
)
from langchain_pinecone import PineconeVectorStore
if os.environ.get("PINECONE_API_KEY", None) is None:
raise Exception("Missing `PINECONE_API_KEY` environment variable.")
if os.environ.get("PINECONE_ENVIRONMENT", None) is None:
raise Exception("Missing `PINECONE_ENVIRONMENT` environment variable.")
PINECONE_INDEX_NAME = os.environ.get("PINECONE_INDEX", "langchain-test")
### Ingest code - you may need to run this the first time
# # Load
# from langchain_community.document_loaders import WebBaseLoader
# loader = WebBaseLoader("https://lilianweng.github.io/posts/2023-06-23-agent/")
# data = loader.load()
# # Split
# from langchain_text_splitters import RecursiveCharacterTextSplitter
# text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)
# all_splits = text_splitter.split_documents(data)
# # Add to vectorDB
# vectorstore = PineconeVectorStore.from_documents(
# documents=all_splits, embedding=OpenAIEmbeddings(), index_name=PINECONE_INDEX_NAME
# )
# retriever = vectorstore.as_retriever()
vectorstore = PineconeVectorStore.from_existing_index(
PINECONE_INDEX_NAME, OpenAIEmbeddings()
)
retriever = vectorstore.as_retriever()
# Condense a chat history and follow-up question into a standalone question
_template = """Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question, in its original language.
Chat History:
{chat_history}
Follow Up Input: {question}
Standalone question:""" # noqa: E501
CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(_template)
# RAG answer synthesis prompt
template = """Answer the question based only on the following context:
<context>
{context}
</context>"""
ANSWER_PROMPT = ChatPromptTemplate.from_messages(
[
("system", template),
MessagesPlaceholder(variable_name="chat_history"),
("user", "{question}"),
]
)
# Conversational Retrieval Chain
DEFAULT_DOCUMENT_PROMPT = PromptTemplate.from_template(template="{page_content}")
def _combine_documents(
docs, document_prompt=DEFAULT_DOCUMENT_PROMPT, document_separator="\n\n"
):
doc_strings = [format_document(doc, document_prompt) for doc in docs]
return document_separator.join(doc_strings)
def _format_chat_history(chat_history: List[Tuple[str, str]]) -> List:
buffer = []
for human, ai in chat_history:
buffer.append(HumanMessage(content=human))
buffer.append(AIMessage(content=ai))
return buffer
# User input
class ChatHistory(BaseModel):
chat_history: List[Tuple[str, str]] = Field(..., extra={"widget": {"type": "chat"}})
question: str
_search_query = RunnableBranch(
# If input includes chat_history, we condense it with the follow-up question
(
RunnableLambda(lambda x: bool(x.get("chat_history"))).with_config(
run_name="HasChatHistoryCheck"
), # Condense follow-up question and chat into a standalone_question
RunnablePassthrough.assign(
chat_history=lambda x: _format_chat_history(x["chat_history"])
)
| CONDENSE_QUESTION_PROMPT
| ChatOpenAI(temperature=0)
| StrOutputParser(),
),
# Else, we have no chat history, so just pass through the question
RunnableLambda(itemgetter("question")),
)
_inputs = RunnableParallel(
{
"question": lambda x: x["question"],
"chat_history": lambda x: _format_chat_history(x["chat_history"]),
"context": _search_query | retriever | _combine_documents,
}
).with_types(input_type=ChatHistory)
chain = _inputs | ANSWER_PROMPT | ChatOpenAI() | StrOutputParser()