langchain学习（四）

Interface官方文档

https://python.langchain.com/docs/expression_language/interfaceTo make it as easy as possible to create custom chains, we’vehttps://python.langchain.com/docs/expression_language/interface

1、langchain采用了一个叫“Runnable” 的标准协议/接口，给用户组合部件（component）形成chains，标准化的接口还包括：

stream：流式（实时）返回结果

invoke：单输入

batch：通过列表进行多输入

同时提供以上接口的异步方式

2、component的输入输出类型表

Component	Input Type	Output Type
Prompt	Dictionary	PromptValue
ChatModel	Single string, list of chat messages or a PromptValue	ChatMessage
LLM	Single string, list of chat messages or a PromptValue	String
OutputParser	The output of an LLM or ChatModel	Depends on the parser
Retriever	Single string	List of Documents
Tool	Single string or dictionary, depending on the tool	Depends on the tool

3、并行请求

方式1通过RunnableParallel

from langchain.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnableParallel
from time import process_time
process_time()
chain1 = ChatPromptTemplate.from_template("tell me a joke about {topic}")
chain2 = (
    ChatPromptTemplate.from_template("write a short (2 line) poem about {topic}")
)
combined = RunnableParallel(joke=chain1, poem=chain2)
print(process_time())

方式2通过batch接口

combined.batch([{"topic": "bears"}, {"topic": "cats"}])