LangChain 41 深入理解LangChain 表达式语言五为什么要用LCEL调用大模型LLM LangChain Expression Language (LCEL)

最新推荐文章于 2024-09-26 09:15:00 发布

AI架构师易筋

最新推荐文章于 2024-09-26 09:15:00 发布

阅读量888

点赞数 21

分类专栏： LLM-Large Language Models 文章标签： langchain prompt LLM chatgpt 语言模型

本文链接：https://blog.csdn.net/zgpeace/article/details/135253672

版权

LLM-Large Language Models 专栏收录该内容

185 篇文章 121 订阅

订阅专栏

LangChain系列文章

在这里插入图片描述

1. LLM instead of chat model LLM而不是聊天模型

如果我们想要使用一个完成端点而不是一个聊天端点：

1.1 LCEL 实现

from langchain.llms import OpenAI
from langchain_core.runnables import RunnablePassthrough
from langchain.prompts import ChatPromptTemplate
from langchain.chat_models import ChatOpenAI
from langchain_core.output_parsers import StrOutputParser
from dotenv import load_dotenv

load_dotenv()

prompt = ChatPromptTemplate.from_template(
    "Tell me a short joke about {topic}"
)
output_parser = StrOutputParser()

llm = OpenAI(model="gpt-3.5-turbo-instruct")
llm_chain = (
    {"topic": RunnablePassthrough()} 
    | prompt
    | llm
    | output_parser
)

response = llm_chain.invoke("ice cream")
print('response >> ', response)

输出

zgpeace@zgpeaces-MacBook-Pro [19:13:50] [~/Workspace/LLM/langchain-llm-app] [develop *]
-> % python LCEL/llm_lcel.py                                                                                                            1 ↵
response >>  

Why did the ice cream go to therapy?

Because it was feeling a little melon-collie.

1.2 不用LCEL实现

from typing import List

import openai
from langchain_core.runnables import RunnablePassthrough
from langchain.prompts import ChatPromptTemplate
from langchain.chat_models import ChatOpenAI
from langchain_core.output_parsers import StrOutputParser
from dotenv import load_dotenv

load_dotenv()

prompt_template = "Tell me a short joke about {topic}"
client = openai.OpenAI()

def call_llm(prompt_value: str) -> str:
    response = client.completions.create(
        model="gpt-3.5-turbo-instruct",
        prompt=prompt_value,
    )
    return response.choices[0].text

def invoke_llm_chain(topic: str) -> str:
    prompt_value = prompt_template.format(topic=topic)
    return call_llm(prompt_value)

response = invoke_llm_chain("ice cream")
print('response >> ', response)

输出

zgpeace@zgpeaces-MacBook-Pro [19:14:52] [~/Workspace/LLM/langchain-llm-app] [develop *]
-> % python LCEL/llm.py 
response >>  
Why did the ice cream go to therapy?
Because it was going through a

2. Different model provider 切换到访问其他大模型比如 Claude-2

2.1 LCEL 实现

from langchain.chat_models import ChatAnthropic

anthropic = ChatAnthropic(model="claude-2")
anthropic_chain = (
    {"topic": RunnablePassthrough()} 
    | prompt 
    | anthropic
    | output_parser
)

anthropic_chain.invoke("ice cream")

2.2 不用LCEL实现

import anthropic

anthropic_template = f"Human:\n\n{prompt_template}\n\nAssistant:"
anthropic_client = anthropic.Anthropic()

def call_anthropic(prompt_value: str) -> str:
    response = anthropic_client.completions.create(
        model="claude-2",
        prompt=prompt_value,
        max_tokens_to_sample=256,
    )
    return response.completion    

def invoke_anthropic_chain(topic: str) -> str:
    prompt_value = anthropic_template.format(topic=topic)
    return call_anthropic(prompt_value)

invoke_anthropic_chain("ice cream")