from langchain.llms.openai import OpenAI

llm = OpenAI()
response = llm.invoke("四大名著指的是什么?")
print(response)

(LLM) ?  language_model python3 00_language_model.py
四大名著指的是《西游記》,、《水滸傳》、《紅樓夢》和《三國演義》,。
(LLM) ?  language_model

from langchain.chat_models import ChatOpenAI
from langchain.schema.messages import HumanMessage, SystemMessage

messages = [
    SystemMessage(content="你一名唐詩鑒賞專家"),
    HumanMessage(content="《蜀道難》"),
]

chat = ChatOpenAI()
print(chat.invoke(messages))

(LLM) ?  language_model python3 00_chat_modes.py
content='《蜀道難》是唐代文學(xué)家李白創(chuàng)作的一首抒發(fā)壯志豪情的長詩。這首詩以描寫艱難險阻的蜀道為背景,，表達(dá)了李白對生活困難的挑戰(zhàn)精神和不屈的意志,。\n\n詩中首先描繪了蜀道的險峻和艱難,，以山峰陡峭,、道路崎嶇為景，展現(xiàn)了蜀道的險阻和危險,，同時也反映了人生道路的曲折困難,。詩中用“青泥何盤盤，百步九折縈巖巒”來形容蜀道的險峻,，這種描寫手法生動地表現(xiàn)了蜀道的險阻之處,。\n\n接著，詩人轉(zhuǎn)而描繪了自己在蜀道上的行進(jìn),，以及在艱難險阻中的堅持和奮斗,。詩人用“路盡燈火落，石墮天壇空”來描述路途的終點(diǎn)和自己的迷茫,，以及用“崖峽千尋翠”來形容自己與險峰相對的壯麗景色,。這種對比描寫，既表現(xiàn)了詩人在困難中的堅持,，也表達(dá)了對自然景色的贊美,。\n\n最后，詩人表達(dá)了自己的豪情壯志和不屈的決心,。他用“昨夜四無鄰,，今宵五陵六扇”來表達(dá)自己的孤獨(dú)和辛苦，以及用“想當(dāng)年,，金戈鐵馬,，氣吞萬里如虎”來表達(dá)自己的豪情壯志。這種豪情壯志的表達(dá),，體現(xiàn)了李白追求自由,、奔放的個性和對人生的樂觀態(tài)度。\n\n《蜀道難》以其雄渾豪情和獨(dú)特的描寫手法,，展現(xiàn)了李白的文學(xué)才華和個人精神,。它通過對自然景色和個人經(jīng)歷的描繪，表達(dá)了對困難的挑戰(zhàn)和對生活的熱愛,，具有濃厚的時代氣息和藝術(shù)魅力,。'
(LLM) ?  language_model

#!/usr/bin/env python3
# -*- coding:utf-8 -*-
from langchain.prompts import PromptTemplate
from langchain.chat_models import ChatOpenAI

llm = ChatOpenAI() 

#### PromptTemplate ####
prompt_template = PromptTemplate.from_template(
    "Tell me a {adjective} joke about {content}"
    )
prompt = prompt_template.format(adjective="funny", content="chickens")
print(prompt)
print(llm.invoke(prompt))

(LLM) ?  prompt_template python3 00.py
Tell me a funny joke about chickens
content='Why did the chicken go to the seance?\n\nTo talk to the other side!'
(LLM) ?  prompt_template

#!/usr/bin/env python3
# -*- coding:utf-8 -*-
from langchain.prompts import ChatPromptTemplate
from langchain.chat_models import ChatOpenAI

llm = ChatOpenAI()
#### ChatPromptTemplate ####
chat_template = ChatPromptTemplate.from_messages(
    [
        ("system", "You are a helpful ai bot, your name is {name}"),
        ("human", "Hello, how are you doing?"),
        ("ai", "I am doing well, thanks!"),
        ("human", "{content}"),
    ]
)
messages = chat_template.format_messages(name="windeal", content="What is your name")
print(messages)
print(llm(messages))

(LLM) ?  prompt_template python3 00.py
[SystemMessage(content='You are a helpful ai bot, your name is windeal'), HumanMessage(content='Hello, how are you doing?'), AIMessage(content='I am doing well, thanks!'), HumanMessage(content='What is your name')]
content='My name is Windeal. How can I assist you today?'
(LLM) ?  prompt_template

class BaseExampleSelector(ABC):
    """Interface for selecting examples to include in prompts."""

    @abstractmethod
    def select_examples(self, input_variables: Dict[str, str]) -> List[dict]:
        """Select which examples to use based on the inputs."""

#!/usr/bin/env python3
# -*- coding:utf-8 -*-

from typing import List
from langchain.output_parsers import PydanticOutputParser
from langchain.prompts import PromptTemplate
from langchain.llms.openai import OpenAI
from langchain.pydantic_v1 import BaseModel, Field, validator

class Book(BaseModel):
    name: str = Field(description="書名")
    author: str = Field(description="作者")

# Here's another example, but with a compound typed field.
class Actor(BaseModel):
    book_list: List[Book] = Field(description="書籍列表")

parser = PydanticOutputParser(pydantic_object=Actor)

prompt = PromptTemplate(
    template="請介紹{subject}相關(guān)內(nèi)容. \n{format_instructions}\n",
    input_variables=["subject"],
    partial_variables={"format_instructions": parser.get_format_instructions()},
)

llm = OpenAI()

_input = prompt.format_prompt(subject="中國四大名著")

output = llm(_input.to_string())

print(parser.parse(output))

(LLM) ?  output_parsers python3 json_parsers.py
book_list=[Book(name='《紅樓夢》', author='曹雪芹'), Book(name='《西游記》', author='吳承恩'), Book(name='《三國演義》', author='羅貫中'), Book(name='《水滸傳》', author='施耐庵')]
(LLM) ?  output_parsers