RunnableBranch: Dynamically route logic based on input | 🦜?🔗 Langchain

基于輸入的動態路由邏輯，通過上一步的輸出選擇下一步操作，允許創建非確定性鏈。路由保證路由間的結構和連貫。

有以下兩種方法執行路由

1、通過RunnableLambda?（推薦該方法）

2、RunnableBranch

from langchain_community.chat_models import ChatOpenAI
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import PromptTemplatemodel=ChatOpenAI(model="gpt-3.5-turbo",temperature=0)
chain = (PromptTemplate.from_template("""根據用戶的問題, 把它歸為關于 `LangChain`, `OpenAI`, or `Other`.不要回答其他字
<question>
{question}
</question>
歸為：
""")| model| StrOutputParser()
)response=chain.invoke({"question": "怎么調用ChatOpenAI"})
print('問題歸類：',response)
##定義LLM
langchain_chain = (PromptTemplate.from_template("""你是一個langchain專家. \開頭用 "langchain有人告訴我"回答問題. \Question: {question}Answer:""")| model
)
openai_chain = (PromptTemplate.from_template("""你是一個OpenAI專家. \開頭用 "OpenAI有人告訴我"回答問題. \Question: {question}Answer:""")| model
)
general_chain = (PromptTemplate.from_template("""回答下面問題:Question: {question}Answer:""")| model
)
def route(info):if "openai" in info["topic"].lower():return openai_chainelif "langchain" in info["topic"].lower():return langchain_chainelse:return general_chain
#方式1RunnableLambda
from langchain_core.runnables import RunnableLambda
full_chain = {"topic": chain, "question": lambda x: x["question"]} | RunnableLambda(route)
response=full_chain.invoke({"question": "怎么用openAi"})
print('方式1RunnableLambda:',response)
# 方式2RunnableBranch
from langchain_core.runnables import RunnableBranch
branch = RunnableBranch((lambda x: "anthropic" in x["topic"].lower(), openai_chain),(lambda x: "langchain" in x["topic"].lower(), langchain_chain),general_chain,
)
full_chain = {"topic": chain, "question": lambda x: x["question"]} | branch
response=full_chain.invoke({"question": "怎么用lanGchaiN?"})
print('方式2RunnableBranch:',response)