icodebroker - 686 중 3 번째 페이지 - 컴퓨터 프로그래밍 코드 데이터베이스

[PYTHON/LANGCHAIN] ChatMessageHistory 클래스 : LLM 채팅하기

■ ChatMessageHistory 클래스를 사용해 LLM과 채팅하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from dotenv                                     import load_dotenv
from langchain_core.prompts                     import ChatPromptTemplate
from langchain_openai                           import ChatOpenAI
from langchain_community.chat_message_histories import ChatMessageHistory

load_dotenv()

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system"     , "You are a helpful assistant. Answer all questions to the best of your ability."),
        ("placeholder", "{messages}")
    ]
)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate | chatOpenAI

chatMessageHistory = ChatMessageHistory()

chatMessageHistory.add_user_message("Translate this sentence from English to French : I love programming.")

responseAIMessage1 = runnableSequence.invoke({"messages" : chatMessageHistory.messages})

print(responseAIMessage1.content)
print("-" * 50)

chatMessageHistory.add_ai_message(responseAIMessage1)

chatMessageHistory.add_user_message("What did I just ask you?")

responseAIMessage2 = runnableSequence.invoke({"messages" : chatMessageHistory.messages})

print(responseAIMessage2.content)
print("-" * 50)

"""
The translation of "I love programming" in French is "J'aime programmer."
--------------------------------------------------
You asked me to translate the sentence "I love programming" from English to French.
--------------------------------------------------
"""

from dotenv import load_dotenv

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

from langchain_community.chat_message_histories import ChatMessageHistory

load_dotenv()

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

("system" , "You are a helpful assistant. Answer all questions to the best of your ability."),

("placeholder", "{messages}")

]

)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate | chatOpenAI

chatMessageHistory = ChatMessageHistory()

chatMessageHistory.add_user_message("Translate this sentence from English to French : I love programming.")

responseAIMessage1 = runnableSequence.invoke({"messages" : chatMessageHistory.messages})

print(responseAIMessage1.content)

print("-" * 50)

chatMessageHistory.add_ai_message(responseAIMessage1)

chatMessageHistory.add_user_message("What did I just ask you?")

responseAIMessage2 = runnableSequence.invoke({"messages" : chatMessageHistory.messages})

print(responseAIMessage2.content)

print("-" * 50)

"""

The translation of "I love programming" in French is "J'aime programmer."

--------------------------------------------------

You asked me to translate the sentence "I love programming" from English to French.

--------------------------------------------------

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.10
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.7.0
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
colorama==0.4.6
dataclasses-json==0.6.7
distro==1.9.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
httpx-sse==0.4.0
idna==3.10
jiter==0.8.0
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.10
langchain-community==0.3.10
langchain-core==0.3.22
langchain-openai==0.2.11
langchain-text-splitters==0.3.2
langsmith==0.1.147
marshmallow==3.23.1
multidict==6.1.0
mypy-extensions==1.0.0
numpy==2.2.0
openai==1.57.0
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic-settings==2.6.1
pydantic_core==2.27.1
python-dotenv==1.0.1
PyYAML==6.0.2
regex==2024.11.6
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
tiktoken==0.8.0
tqdm==4.67.1
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.10

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.7.0

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

colorama==0.4.6

dataclasses-json==0.6.7

distro==1.9.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

httpx-sse==0.4.0

idna==3.10

jiter==0.8.0

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.10

langchain-community==0.3.10

langchain-core==0.3.22

langchain-openai==0.2.11

langchain-text-splitters==0.3.2

langsmith==0.1.147

marshmallow==3.23.1

multidict==6.1.0

mypy-extensions==1.0.0

numpy==2.2.0

openai==1.57.0

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic-settings==2.6.1

pydantic_core==2.27.1

python-dotenv==1.0.1

PyYAML==6.0.2

regex==2024.11.6

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

tiktoken==0.8.0

tqdm==4.67.1

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

[PYTHON/LANGCHAIN] ChatMessageHistory 클래스 : add_user_message/add_ai_message 메소드를 사용해 메시지 추가하기

■ ChatMessageHistory 클래스의 add_user_message/add_ai_message 메소드를 사용해 메시지를 추가하는 방법을 보여준다. ▶ main.py


from langchain_community.chat_message_histories import ChatMessageHistory

chatMessageHistory = ChatMessageHistory()

chatMessageHistory.add_user_message("Translate this sentence from English to French : I love programming.")
chatMessageHistory.add_ai_message("J'adore la programmation.")

messageList = chatMessageHistory.messages

print(messageList)

"""
[
    HumanMessage(
        content           = 'Translate this sentence from English to French : I love programming.',
        additional_kwargs = {},
        response_metadata = {}
    ),
    AIMessage(
        content           = "J'adore la programmation.",
        additional_kwargs = {},
        response_metadata = {}
    )
]
"""

from langchain_community.chat_message_histories import ChatMessageHistory

chatMessageHistory = ChatMessageHistory()

chatMessageHistory.add_user_message("Translate this sentence from English to French : I love programming.")

chatMessageHistory.add_ai_message("J'adore la programmation.")

messageList = chatMessageHistory.messages

print(messageList)

"""

[

HumanMessage(

content = 'Translate this sentence from English to French : I love programming.',

additional_kwargs = {},

response_metadata = {}

AIMessage(

content = "J'adore la programmation.",

additional_kwargs = {},

response_metadata = {}

)

]

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.10
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.7.0
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
dataclasses-json==0.6.7
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
httpx-sse==0.4.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.10
langchain-community==0.3.10
langchain-core==0.3.22
langchain-text-splitters==0.3.2
langsmith==0.1.147
marshmallow==3.23.1
multidict==6.1.0
mypy-extensions==1.0.0
numpy==2.2.0
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic-settings==2.6.1
pydantic_core==2.27.1
python-dotenv==1.0.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.10

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.7.0

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

dataclasses-json==0.6.7

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

httpx-sse==0.4.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.10

langchain-community==0.3.10

langchain-core==0.3.22

langchain-text-splitters==0.3.2

langsmith==0.1.147

marshmallow==3.23.1

multidict==6.1.0

mypy-extensions==1.0.0

numpy==2.2.0

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic-settings==2.6.1

pydantic_core==2.27.1

python-dotenv==1.0.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain-community 명령을 실행했다.

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : invoke 메소드에서 MessagePlaceHloder 객체에 인자 전달하기

■ RunnableSequence 클래스의 invoke 메소드에서 MessagePlaceHloder 객체에 인자를 전달하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from dotenv                 import load_dotenv
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai       import ChatOpenAI

load_dotenv()

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system"     , "You are a helpful assistant. Answer all questions to the best of your ability."),
        ("placeholder", "{messages}")
    ]
)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate | chatOpenAI

responseAIMessage = runnableSequence.invoke(
    {
        "messages" : [
            ("human", "Translate this sentence from English to French : I love programming."),
            ("ai"   , "J'adore la programmation."                                           ),
            ("human", "What did you just say?"                                              )
        ]
    }
)

print(responseAIMessage.content)

"""
I translated the sentence "I love programming" into French, which is "J'adore la programmation."
"""

from dotenv import load_dotenv

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

load_dotenv()

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

("system" , "You are a helpful assistant. Answer all questions to the best of your ability."),

("placeholder", "{messages}")

]

)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate | chatOpenAI

responseAIMessage = runnableSequence.invoke(

{

"messages" : [

("human", "Translate this sentence from English to French : I love programming."),

("ai" , "J'adore la programmation." ),

("human", "What did you just say?" )

]

}

)

print(responseAIMessage.content)

"""

I translated the sentence "I love programming" into French, which is "J'adore la programmation."

"""

[PYTHON/LANGCHAIN] ChatPromptTemplate 클래스 : from_messages 정적 메소드를 사용해 암시적으로 MessagesPlaceholder 객체 만들기

■ ChatPromptTemplate 클래스의 from_messages 정적 메소드를 사용해 암시적으로 MessagesPlaceholder 객체를 만드는 방법을 보여준다. ▶ main.py


from langchain_core.prompts import ChatPromptTemplate

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system"     , "You are a helpful assistant. Answer all questions to the best of your ability."),
        ("placeholder", "{messages}")
    ]
)

print(chatPromptTemplate)

"""
input_variables    = []
optional_variables = ['messages']
input_types        = {
    'messages' : list[
        typing.Annotated[
            typing.Union[
                typing.Annotated[langchain_core.messages.ai.AIMessage                 , Tag(tag = 'ai'                  )],
                typing.Annotated[langchain_core.messages.human.HumanMessage           , Tag(tag = 'human'               )],
                typing.Annotated[langchain_core.messages.chat.ChatMessage             , Tag(tag = 'chat'                )],
                typing.Annotated[langchain_core.messages.system.SystemMessage         , Tag(tag = 'system'              )],
                typing.Annotated[langchain_core.messages.function.FunctionMessage     , Tag(tag = 'function'            )],
                typing.Annotated[langchain_core.messages.tool.ToolMessage             , Tag(tag = 'tool'                )],
                typing.Annotated[langchain_core.messages.ai.AIMessageChunk            , Tag(tag = 'AIMessageChunk'      )],
                typing.Annotated[langchain_core.messages.human.HumanMessageChunk      , Tag(tag = 'HumanMessageChunk'   )],
                typing.Annotated[langchain_core.messages.chat.ChatMessageChunk        , Tag(tag = 'ChatMessageChunk'    )],
                typing.Annotated[langchain_core.messages.system.SystemMessageChunk    , Tag(tag = 'SystemMessageChunk'  )],
                typing.Annotated[langchain_core.messages.function.FunctionMessageChunk, Tag(tag = 'FunctionMessageChunk')],
                typing.Annotated[langchain_core.messages.tool.ToolMessageChunk        , Tag(tag = 'ToolMessageChunk'    )]
            ],
            FieldInfo(
                annotation    = NoneType,
                required      = True,
                discriminator = Discriminator(
                    discriminator        = <function _get_type at 0x0000024FD2ACA480>,
                    custom_error_type    = None,
                    custom_error_message = None,
                    custom_error_context = None
                )
            )
        ]
    ]
}
partial_variables = {'messages' : []}
messages          = [
    SystemMessagePromptTemplate(
        prompt = PromptTemplate(
            input_variables   = [],
            input_types       = {},
            partial_variables = {},
            template          = 'You are a helpful assistant. Answer all questions to the best of your ability.'
        ),
        additional_kwargs = {}
    ),
    MessagesPlaceholder(variable_name = 'messages', optional = True)
]
"""

from langchain_core.prompts import ChatPromptTemplate

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

("system" , "You are a helpful assistant. Answer all questions to the best of your ability."),

("placeholder", "{messages}")

]

)

print(chatPromptTemplate)

"""

input_variables = []

optional_variables = ['messages']

input_types = {

'messages' : list[

typing.Annotated[

typing.Union[

typing.Annotated[langchain_core.messages.ai.AIMessage , Tag(tag = 'ai' )],

typing.Annotated[langchain_core.messages.human.HumanMessage , Tag(tag = 'human' )],

typing.Annotated[langchain_core.messages.chat.ChatMessage , Tag(tag = 'chat' )],

typing.Annotated[langchain_core.messages.system.SystemMessage , Tag(tag = 'system' )],

typing.Annotated[langchain_core.messages.function.FunctionMessage , Tag(tag = 'function' )],

typing.Annotated[langchain_core.messages.tool.ToolMessage , Tag(tag = 'tool' )],

typing.Annotated[langchain_core.messages.ai.AIMessageChunk , Tag(tag = 'AIMessageChunk' )],

typing.Annotated[langchain_core.messages.human.HumanMessageChunk , Tag(tag = 'HumanMessageChunk' )],

typing.Annotated[langchain_core.messages.chat.ChatMessageChunk , Tag(tag = 'ChatMessageChunk' )],

typing.Annotated[langchain_core.messages.system.SystemMessageChunk , Tag(tag = 'SystemMessageChunk' )],

typing.Annotated[langchain_core.messages.function.FunctionMessageChunk, Tag(tag = 'FunctionMessageChunk')],

typing.Annotated[langchain_core.messages.tool.ToolMessageChunk , Tag(tag = 'ToolMessageChunk' )]

FieldInfo(

annotation = NoneType,

required = True,

discriminator = Discriminator(

discriminator = <function _get_type at 0x0000024FD2ACA480>,

custom_error_type = None,

custom_error_message = None,

custom_error_context = None

)

]

}

partial_variables = {'messages' : []}

messages = [

SystemMessagePromptTemplate(

prompt = PromptTemplate(

input_variables = [],

input_types = {},

partial_variables = {},

template = 'You are a helpful assistant. Answer all questions to the best of your ability.'

additional_kwargs = {}

MessagesPlaceholder(variable_name = 'messages', optional = True)

]

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.10
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.7.0
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.10
langchain-core==0.3.22
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==2.2.0
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.10

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.7.0

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.10

langchain-core==0.3.22

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==2.2.0

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install

[PYTHON/LANGCHAIN] 커스텀 출력 파서 만들기 2

■ 커스텀 출력 파서를 만드는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


import re
import json

from dotenv                  import load_dotenv
from pydantic                import BaseModel
from pydantic                import Field
from typing                  import List
from langchain_core.prompts  import ChatPromptTemplate
from langchain_openai        import ChatOpenAI
from langchain_core.messages import AIMessage

load_dotenv()

class Person(BaseModel):
    """Information about a person."""

    name             : str   = Field(..., description = "The name of the person"                       )
    height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):
    """Identifying information about all people in a text."""

    people : List[Person]

chatPromptTemplate1 = ChatPromptTemplate.from_messages(
    [
        ("system", "Answer the user query. Output your answer as JSON that matches the given schema: ```json\n{schema}\n```. Make sure to wrap the answer in ```json and ``` tags"),
        ("human" , "{query}"),
    ]
)

chatPromptTemplate2 = chatPromptTemplate1.partial(schema = People.model_json_schema())

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

def extractJSON(aiMessage : AIMessage) -> List[dict]:
    """Extracts JSON content from a string where JSON is embedded between ```json and ``` tags.

    Parameters :
        text (str) : The text containing the JSON content.

    Returns :
        list : A list of extracted JSON strings.
    """
    text = aiMessage.content
    # JSON 블록과 일치하도록 정규 표현식 패턴을 정의한다.
    pattern = r"```json(.*?)```"
    # 문자열에서 패턴의 겹치지 않는 모든 일치 항목을 찾는다.
    matcheStringList = re.findall(pattern, text, re.DOTALL)
    # 앞이나 뒤의 공백을 제거하고 일치하는 JSON 문자열 목록을 반환한다.
    try:
        return [json.loads(matchString.strip()) for matchString in matcheStringList]
    except Exception:
        raise ValueError(f"Failed to parse: {aiMessage}")

runnableSequence = chatPromptTemplate2 | chatOpenAI | extractJSON

responsePeopleList = runnableSequence.invoke({"query" : "Anna is 23 years old and she is 6 feet tall"})

print(responsePeopleList)

"""
[{'people': [{'name': 'Anna', 'height_in_meters': 1.8288}]}]
"""

import re

import json

from dotenv import load_dotenv

from pydantic import BaseModel

from pydantic import Field

from typing import List

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

from langchain_core.messages import AIMessage

load_dotenv()

class Person(BaseModel):

"""Information about a person."""

name : str = Field(..., description = "The name of the person" )

height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):

"""Identifying information about all people in a text."""

people : List[Person]

chatPromptTemplate1 = ChatPromptTemplate.from_messages(

[

("system", "Answer the user query. Output your answer as JSON that matches the given schema: ```json\n{schema}\n```. Make sure to wrap the answer in ```json and ``` tags"),

("human" , "{query}"),

]

)

chatPromptTemplate2 = chatPromptTemplate1.partial(schema = People.model_json_schema())

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

def extractJSON(aiMessage : AIMessage) -> List[dict]:

"""Extracts JSON content from a string where JSON is embedded between ```json and ``` tags.

Parameters :

text (str) : The text containing the JSON content.

Returns :

list : A list of extracted JSON strings.

"""

text = aiMessage.content

# JSON 블록과 일치하도록 정규 표현식 패턴을 정의한다.

pattern = r"```json(.*?)```"

# 문자열에서 패턴의 겹치지 않는 모든 일치 항목을 찾는다.

matcheStringList = re.findall(pattern, text, re.DOTALL)

# 앞이나 뒤의 공백을 제거하고 일치하는 JSON 문자열 목록을 반환한다.

try:

return [json.loads(matchString.strip()) for matchString in matcheStringList]

except Exception:

raise ValueError(f"Failed to parse: {aiMessage}")

runnableSequence = chatPromptTemplate2 | chatOpenAI | extractJSON

responsePeopleList = runnableSequence.invoke({"query" : "Anna is 23 years old and she is 6 feet tall"})

print(responsePeopleList)

"""

[{'people': [{'name': 'Anna', 'height_in_meters': 1.8288}]}]

"""

▶ requirements.txt


annotated-types==0.7.0
anyio==4.7.0
certifi==2024.8.30
charset-normalizer==3.4.0
colorama==0.4.6
distro==1.9.0
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
idna==3.10
jiter==0.8.0
jsonpatch==1.33
jsonpointer==3.0.0
langchain-core==0.3.22
langchain-openai==0.2.11
langsmith==0.1.147
openai==1.57.0
orjson==3.10.12
packaging==24.2
pydantic==2.10.3
pydantic_core==2.27.1
python-dotenv==1.0.1
PyYAML==6.0.2
regex==2024.11.6
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
tenacity==9.0.0
tiktoken==0.8.0
tqdm==4.67.1
typing_extensions==4.12.2
urllib3==2.2.3

annotated-types==0.7.0

anyio==4.7.0

certifi==2024.8.30

charset-normalizer==3.4.0

colorama==0.4.6

distro==1.9.0

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

idna==3.10

jiter==0.8.0

jsonpatch==1.33

jsonpointer==3.0.0

langchain-core==0.3.22

langchain-openai==0.2.11

langsmith==0.1.147

openai==1.57.0

orjson==3.10.12

packaging==24.2

pydantic==2.10.3

pydantic_core==2.27.1

python-dotenv==1.0.1

PyYAML==6.0.2

regex==2024.11.6

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

tenacity==9.0.0

tiktoken==0.8.0

tqdm==4.67.1

typing_extensions==4.12.2

urllib3==2.2.3

※

[PYTHON/LANGCHAIN] PydanticOutputParser 클래스 : 정형화된 데이터 추출하기 (프롬프트만 사용)

■ PydanticOutputParser 클래스를 사용해 정형화된 데이터를 추출하는 방법을 보여준다. (프롬프트만 사용) ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from dotenv                        import load_dotenv
from pydantic                      import BaseModel
from pydantic                      import Field
from typing                        import List
from langchain_core.output_parsers import PydanticOutputParser
from langchain_core.prompts        import ChatPromptTemplate
from langchain_openai              import ChatOpenAI

load_dotenv()

class Person(BaseModel):
    """Information about a person."""

    name             : str   = Field(..., description = "The name of the person"                       )
    height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):
    """Identifying information about all people in a text."""

    people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

chatPromptTemplate1 = ChatPromptTemplate.from_messages(
    [
        ("system", "Answer the user query. Wrap the output in `json` tags\n{format_instructions}"),
        ("human" , "{query}")
    ]
)

chatPromptTemplate2 = chatPromptTemplate1.partial(format_instructions = pydanticOutputParser.get_format_instructions())

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate2 | chatOpenAI | pydanticOutputParser

responsePeople = runnableSequence.invoke({"query" : "Anna is 23 years old and she is 6 feet tall"})

print(responsePeople)

"""
people=[Person(name='Anna', height_in_meters=1.8288)]
"""

from dotenv import load_dotenv

from pydantic import BaseModel

from pydantic import Field

from typing import List

from langchain_core.output_parsers import PydanticOutputParser

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

load_dotenv()

class Person(BaseModel):

"""Information about a person."""

name : str = Field(..., description = "The name of the person" )

height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):

"""Identifying information about all people in a text."""

people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

chatPromptTemplate1 = ChatPromptTemplate.from_messages(

[

("system", "Answer the user query. Wrap the output in `json` tags\n{format_instructions}"),

("human" , "{query}")

]

)

chatPromptTemplate2 = chatPromptTemplate1.partial(format_instructions = pydanticOutputParser.get_format_instructions())

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini")

runnableSequence = chatPromptTemplate2 | chatOpenAI | pydanticOutputParser

responsePeople = runnableSequence.invoke({"query" : "Anna is 23 years old and she is 6 feet tall"})

print(responsePeople)

"""

people=[Person(name='Anna', height_in_meters=1.8288)]

"""

[PYTHON/LANGCHAIN] ChatPromptTemplate 클래스 : partial 메소드를 사용해 매개 변수 값을 설정하고 새로운 ChatPromptTemplate 객체 만들기

■ ChatPromptTemplate 클래스의 partial 메소드를 사용해 매개 변수 값을 설정하고 새로운 ChatPromptTemplate 객체를 만드는 방법을 보여준다. ▶ main.py


from pydantic                      import BaseModel
from pydantic                      import Field
from typing                        import List
from langchain_core.output_parsers import PydanticOutputParser
from langchain_core.prompts        import ChatPromptTemplate

class Person(BaseModel):
    """Information about a person."""

    name             : str   = Field(..., description = "The name of the person"                       )
    height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):
    """Identifying information about all people in a text."""

    people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

chatPromptTemplate1 = ChatPromptTemplate.from_messages(
    [
        ("system", "Answer the user query. Wrap the output in `json` tags\n{format_instructions}"),
        ("human" , "{query}")
    ]
)

chatPromptTemplate2 = chatPromptTemplate1.partial(format_instructions = pydanticOutputParser.get_format_instructions())

from pydantic import BaseModel

from pydantic import Field

from typing import List

from langchain_core.output_parsers import PydanticOutputParser

from langchain_core.prompts import ChatPromptTemplate

class Person(BaseModel):

"""Information about a person."""

name : str = Field(..., description = "The name of the person" )

height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):

"""Identifying information about all people in a text."""

people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

chatPromptTemplate1 = ChatPromptTemplate.from_messages(

[

("system", "Answer the user query. Wrap the output in `json` tags\n{format_instructions}"),

("human" , "{query}")

]

)

chatPromptTemplate2 = chatPromptTemplate1.partial(format_instructions = pydanticOutputParser.get_format_instructions())

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.10
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.7.0
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.10
langchain-core==0.3.22
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==2.1.3
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.10

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.7.0

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.10

langchain-core==0.3.22

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==2.1.3

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

[PYTHON/LANGCHAIN] PydanticOutputParser 클래스 : 생성자에서 pydantic_object 인자를 사용해 PydanticOutputParser 객체 만들기

■ PydanticOutputParser 클래스의 생성자에서 pydantic_object 인자를 사용해 PydanticOutputParser 객체를 만드는 방법을 보여준다. ▶ main.py


from pydantic                      import BaseModel
from pydantic                      import Field
from typing                        import List
from langchain_core.output_parsers import PydanticOutputParser

class Person(BaseModel):
    """Information about a person."""

    name             : str   = Field(..., description = "The name of the person"                       )
    height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):
    """Identifying information about all people in a text."""

    people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

from pydantic import BaseModel

from pydantic import Field

from typing import List

from langchain_core.output_parsers import PydanticOutputParser

class Person(BaseModel):

"""Information about a person."""

name : str = Field(..., description = "The name of the person" )

height_in_meters : float = Field(..., description = "The height of the person expressed in meters.")

class People(BaseModel):

"""Identifying information about all people in a text."""

people : List[Person]

pydanticOutputParser = PydanticOutputParser(pydantic_object = People)

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.10
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.7.0
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.1
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.10
langchain-core==0.3.22
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==2.1.3
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.10

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.7.0

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.1

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.10

langchain-core==0.3.22

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==2.1.3

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain

[PYTHON/PANDAS] Series 클래스 : iloc 속성과 [] 연산자를 사용해 슬라이싱 처리하기

■ Series 클래스의 iloc 속성과 [] 연산자를 사용해 슬라이싱을 처리하는 방법을 보여준다. ▶ main.py


import numpy  as np
import pandas as pd

series1 = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])

print(series1)
print("-" * 50)

"""
a    0.794678
b    1.756467
c   -0.014899
d    0.926113
e   -1.238260
dtype: float64
"""

value1 = series1.iloc[0]

print(value1) # numpy.float64
print("-" * 50)

"""
0.7946776583297425
"""

series2 = series1.iloc[:3]

print(series2)
print("-" * 50)

"""
a    0.794678
b    1.756467
c   -0.014899
dtype: float64
"""

medianValue = series1.median()

print(medianValue)
print("-" * 50)

"""
0.7946776583297425
"""

series3 = series1[series1 > medianValue]

print(series3)
print("-" * 50)

"""
b    1.756467
d    0.926113
dtype: float64
"""

series4 = series1.iloc[[4, 3, 1]]

print(series4)
print("-" * 50)

"""
e   -1.238260
d    0.926113
b    1.756467
dtype: float64
"""

series5 = np.exp(series1) # e^x 연산을 실행한다.

print(series5)
print("-" * 50)

"""
a    2.213727
b    5.791940
c    0.985212
d    2.524677
e    0.289888
dtype: float64
"""

import numpy as np

import pandas as pd

series1 = pd.Series(np.random.randn(5), index=["a", "b", "c", "d", "e"])

print(series1)

print("-" * 50)

"""

a 0.794678

b 1.756467

c -0.014899

d 0.926113

e -1.238260

dtype: float64

"""

value1 = series1.iloc[0]

print(value1) # numpy.float64

print("-" * 50)

"""

0.7946776583297425

"""

series2 = series1.iloc[:3]

print(series2)

print("-" * 50)

"""

a 0.794678

b 1.756467

c -0.014899

dtype: float64

"""

medianValue = series1.median()

print(medianValue)

print("-" * 50)

"""

0.7946776583297425

"""

series3 = series1[series1 > medianValue]

print(series3)

print("-" * 50)

"""

b 1.756467

d 0.926113

dtype: float64

"""

series4 = series1.iloc[[4, 3, 1]]

print(series4)

print("-" * 50)

"""

e -1.238260

d 0.926113

b 1.756467

dtype: float64

"""

series5 = np.exp(series1) # e^x 연산을 실행한다.

print(series5)

print("-" * 50)

"""

a 2.213727

b 5.791940

c 0.985212

d 2.524677

e 0.289888

dtype: float64

"""

▶ requirements.txt


numpy==2.1.3
pandas==2.2.3
python-dateutil==2.9.0.post0
pytz==2024.2
six==1.17.0
tzdata==2024.2

numpy==2.1.3

pandas==2.2.3

python-dateutil==2.9.0.post0

pytz==2024.2

six==1.17.0

tzdata==2024.2

※ pip install pandas

[PYTHON/PANDAS] Series 클래스 : 생성자에서 index 인자와 dict 객체를 사용해 Series 객체 만들기

■ Series 클래스의 생성자에서 index 인자와 dict 객체를 사용해 Series 객체를 만드는 방법을 보여준다. ▶ main.py


import pandas as pd

dictionary1 = {"b" : 1, "a" : 0, "c" : 2}

series1 = pd.Series(dictionary1)

print(series1)
print()

"""
b    1
a    0
c    2
dtype: int64
"""

dictionary2 = {"a" : 0.0, "b" : 1.0, "c" : 2.0}

series2 = pd.Series(dictionary2)

print(series2)
print()

"""
a    0.0
b    1.0
c    2.0
dtype: float64
"""

series3 = pd.Series(dictionary2, index = ["b", "c", "d", "a"])

print(series3)
print("-" *50)

""""
b    1.0
c    2.0
d    NaN
a    0.0
dtype: float64
"""

import pandas as pd

dictionary1 = {"b" : 1, "a" : 0, "c" : 2}

series1 = pd.Series(dictionary1)

print(series1)

print()

"""

b 1

a 0

c 2

dtype: int64

"""

dictionary2 = {"a" : 0.0, "b" : 1.0, "c" : 2.0}

series2 = pd.Series(dictionary2)

print(series2)

print()

"""

a 0.0

b 1.0

c 2.0

dtype: float64

"""

series3 = pd.Series(dictionary2, index = ["b", "c", "d", "a"])

print(series3)

print("-" *50)

""""

b 1.0

c 2.0

d NaN

a 0.0

dtype: float64

"""

▶ requirements.txt


numpy==2.1.3
pandas==2.2.3
python-dateutil==2.9.0.post0
pytz==2024.2
six==1.17.0
tzdata==2024.2

numpy==2.1.3

pandas==2.2.3

python-dateutil==2.9.0.post0

pytz==2024.2

six==1.17.0

tzdata==2024.2

※ pip

[PYTHON/PANDAS] Series 클래스 : index 속성 사용하기

■ Series 클래스의 index 속성을 사용하는 방법을 보여준다. ▶ main.py


import numpy  as np
import pandas as pd

series = pd.Series(np.random.randn(5), index = ["a", "b", "c", "d", "e"])

print(series)

"""
a    0.229116
b    0.184222
c    1.208261
d    1.945010
e    1.371544
dtype: float64
"""

index = series.index

print(index) # pandas.core.indexes.base.Index

"""
Index(['a', 'b', 'c', 'd', 'e'], dtype='object')
"""

import numpy as np

import pandas as pd

series = pd.Series(np.random.randn(5), index = ["a", "b", "c", "d", "e"])

print(series)

"""

a 0.229116

b 0.184222

c 1.208261

d 1.945010

e 1.371544

dtype: float64

"""

index = series.index

print(index) # pandas.core.indexes.base.Index

"""

Index(['a', 'b', 'c', 'd', 'e'], dtype='object')

"""

▶ requirements.txt


numpy==2.1.3
pandas==2.2.3
python-dateutil==2.9.0.post0
pytz==2024.2
six==1.17.0
tzdata==2024.2

numpy==2.1.3

pandas==2.2.3

python-dateutil==2.9.0.post0

pytz==2024.2

six==1.17.0

tzdata==2024.2

※ pip install pandas 명령을 실행했다.

[PYTHON/PANDAS] Series 클래스 : 생성자에서 index 인자와 NDArray 객체를 사용해 Series 객체 만들기

■ Series 클래스의 생성자에서 index 인자와 NDArray 객체를 사용해 Series 객체를 만드는 방법을 보여준다. ▶ main.py


import numpy  as np
import pandas as pd

series = pd.Series(np.random.randn(5), index = ["a", "b", "c", "d", "e"])

print(series)

"""
a    0.229116
b    0.184222
c    1.208261
d    1.945010
e    1.371544
dtype: float64
"""

import numpy as np

import pandas as pd

series = pd.Series(np.random.randn(5), index = ["a", "b", "c", "d", "e"])

print(series)

"""

a 0.229116

b 0.184222

c 1.208261

d 1.945010

e 1.371544

dtype: float64

"""

▶ requirements.txt


numpy==2.1.3
pandas==2.2.3
python-dateutil==2.9.0.post0
pytz==2024.2
six==1.17.0
tzdata==2024.2

numpy==2.1.3

pandas==2.2.3

python-dateutil==2.9.0.post0

pytz==2024.2

six==1.17.0

tzdata==2024.2

※ pip

[TOOL/VS CODE] 윈도우즈 운영 체제에서 명령 프롬프트를 디폴트 터미널로 설정하기

■ 윈도우즈 운영 체제에서 명령 프롬프트를 디폴트 터미널로 설정하는 방법을 보여준다. 1. [Visual Studio Code]를 실행한다. 2. CTRL + , 키를 눌러서

[PYTHON/LANGCHAIN] RunnableParallel 클래스 : invoke 메소드 사용하기 2

■ RunnableParallel 클래스의 invoke 메소드를 사용하는 방법을 보여준다. ▶ main.py


from langchain_core.runnables import RunnableParallel
from langchain_core.runnables import RunnablePassthrough

runnableParallel = RunnableParallel(
    original = RunnablePassthrough(),
    key      = lambda x : x["key"  ],
    value    = lambda x : x["value"]
)

responseDictionary = runnableParallel.invoke({"key" : "name", "value" : "david"})

print(responseDictionary)

"""
{'original': {'key': 'name', 'value': 'david'}, 'key': 'name', 'value': 'david'}
"""

from langchain_core.runnables import RunnableParallel

from langchain_core.runnables import RunnablePassthrough

runnableParallel = RunnableParallel(

original = RunnablePassthrough(),

key = lambda x : x["key" ],

value = lambda x : x["value"]

)

responseDictionary = runnableParallel.invoke({"key" : "name", "value" : "david"})

print(responseDictionary)

"""

{'original': {'key': 'name', 'value': 'david'}, 'key': 'name', 'value': 'david'}

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
async-timeout==4.0.3
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
exceptiongroup==1.2.2
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==1.26.4
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

async-timeout==4.0.3

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

exceptiongroup==1.2.2

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==1.26.4

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain 명령을 실행했다.

[PYTHON/LANGCHAIN] RunnableParallel 클래스 : invoke 메소드 사용하기 1

■ RunnableParallel 클래스의 invoke 메소드를 사용하는 방법을 보여준다. ▶ main.py


from langchain_core.runnables import RunnableParallel
from langchain_core.runnables import RunnablePassthrough

runnableParallel = RunnableParallel(
    original  = RunnablePassthrough(),
    uppercase = lambda x : x.upper(),
    length    = lambda x : len(x)
)

responseDictionary = runnableParallel.invoke("Hello")

print(responseDictionary)

"""
{'original': 'Hello', 'uppercase': 'HELLO', 'length': 5}
"""

from langchain_core.runnables import RunnableParallel

from langchain_core.runnables import RunnablePassthrough

runnableParallel = RunnableParallel(

original = RunnablePassthrough(),

uppercase = lambda x : x.upper(),

length = lambda x : len(x)

)

responseDictionary = runnableParallel.invoke("Hello")

print(responseDictionary)

"""

{'original': 'Hello', 'uppercase': 'HELLO', 'length': 5}

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
async-timeout==4.0.3
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
exceptiongroup==1.2.2
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==1.26.4
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

async-timeout==4.0.3

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

exceptiongroup==1.2.2

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==1.26.4

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain 명령을 실행했다.

[PYTHON/LANGCHAIN] RunnablePassthrough 클래스 : invoke 메소드 사용하기

■ RunnablePassthrough 클래스의 invoke 메소드를 사용하는 방법을 보여준다. ▶ main.py


from langchain_core.runnables import RunnablePassthrough

runnablePassthrough = RunnablePassthrough()

responseString = runnablePassthrough.invoke("Hello")

print(responseString)

"""
Hello
"""

from langchain_core.runnables import RunnablePassthrough

runnablePassthrough = RunnablePassthrough()

responseString = runnablePassthrough.invoke("Hello")

print(responseString)

"""

Hello

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
async-timeout==4.0.3
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
exceptiongroup==1.2.2
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==1.26.4
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

async-timeout==4.0.3

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

exceptiongroup==1.2.2

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==1.26.4

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain 명령을 실행했다.

[PYTHON/LANGCHAIN] 우분투 운영 체제에서 긴 텍스트에서 정형화된 데이터 추출하기 (RAG 기반 접근 방식)

■ 우분투 운영 체제에서 긴 텍스트에서 정형화된 데이터를 추출하는 방법을 보여준다. (RAG 기반 접근 방식) ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에

[PYTHON/LANGCHAIN] 우분투 운영 체제에서 긴 텍스트에서 정형화된 데이터 추출하기 (무차별 대입 방식)

■ 우분투 운영 체제에서 긴 텍스트에서 정형화된 데이터를 추출하는 방법을 보여준다. (무차별 대입 방식) ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다.

[PYTHON/LANGCHAIN] 윈도우즈 운영 체제에서 긴 텍스트에서 정형화된 데이터 추출하기 (무차별 대입 방식)

■ 윈도우즈 운영 체제에서 긴 텍스트에서 정형화된 데이터를 추출하는 방법을 보여준다. (무차별 대입 방식) ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다.

[PYTHON/LANGCHAIN] BSHTMLLoader 클래스 : 윈도우즈 운영 체제에서 UTF-8 인코딩 파일 사용하기

■ BSHTMLLoader 클래스를 사용해 윈도우즈 운영 체제에서 UTF-8 인코딩 파일을 사용하는 방법을 보여준다. ▶ main.py


import requests

from langchain_community.document_loaders import BSHTMLLoader
from langchain.schema                     import Document
from bs4                                  import BeautifulSoup

response = requests.get("https://en.wikipedia.org/wiki/Car")

with open("car.html", "w", encoding = "utf-8") as textIOWrapper:
    textIOWrapper.write(response.text)

class CustomBSHTMLLoader(BSHTMLLoader):
    def load(self) -> list[Document]:
        try:
            with open(self.file_path, "r", encoding = "utf-8") as textIOWrapper:
                beautifulSoup = BeautifulSoup(textIOWrapper, "html.parser")
                text = beautifulSoup.get_text()
        except UnicodeDecodeError:
            print("UTF-8 디코딩 중 오류가 발생했습니다.")
            with open(self.file_path, "r", encoding = "cp949") as textIOWrapper:
                beautifulSoup = BeautifulSoup(textIOWrapper, "html.parser")
                text = beautifulSoup.get_text()
        metadata = {"source" : self.file_path}
        return [Document(page_content = text, metadata = metadata)]

customBSHTMLLoader = CustomBSHTMLLoader("car.html")

documentList = customBSHTMLLoader.load()

import requests

from langchain_community.document_loaders import BSHTMLLoader

from langchain.schema import Document

from bs4 import BeautifulSoup

response = requests.get("https://en.wikipedia.org/wiki/Car")

with open("car.html", "w", encoding = "utf-8") as textIOWrapper:

textIOWrapper.write(response.text)

class CustomBSHTMLLoader(BSHTMLLoader):

def load(self) -> list[Document]:

try:

with open(self.file_path, "r", encoding = "utf-8") as textIOWrapper:

beautifulSoup = BeautifulSoup(textIOWrapper, "html.parser")

text = beautifulSoup.get_text()

except UnicodeDecodeError:

print("UTF-8 디코딩 중 오류가 발생했습니다.")

with open(self.file_path, "r", encoding = "cp949") as textIOWrapper:

beautifulSoup = BeautifulSoup(textIOWrapper, "html.parser")

text = beautifulSoup.get_text()

metadata = {"source" : self.file_path}

return [Document(page_content = text, metadata = metadata)]

customBSHTMLLoader = CustomBSHTMLLoader("car.html")

documentList = customBSHTMLLoader.load()

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
attrs==24.2.0
beautifulsoup4==4.12.3
bs4==0.0.2
certifi==2024.8.30
charset-normalizer==3.4.0
dataclasses-json==0.6.7
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
httpx-sse==0.4.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-community==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
lxml==5.3.0
marshmallow==3.23.1
multidict==6.1.0
mypy-extensions==1.0.0
numpy==2.1.3
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic-settings==2.6.1
pydantic_core==2.27.1
python-dotenv==1.0.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
soupsieve==2.6
SQLAlchemy==2.0.36
tenacity==9.0.0
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

attrs==24.2.0

beautifulsoup4==4.12.3

bs4==0.0.2

certifi==2024.8.30

charset-normalizer==3.4.0

dataclasses-json==0.6.7

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

httpx-sse==0.4.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-community==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

lxml==5.3.0

marshmallow==3.23.1

multidict==6.1.0

mypy-extensions==1.0.0

numpy==2.1.3

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic-settings==2.6.1

pydantic_core==2.27.1

python-dotenv==1.0.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

soupsieve==2.6

SQLAlchemy==2.0.36

tenacity==9.0.0

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install

[PYTHON/LANGCHAIN] 정형화된 데이터 추출시 참조 예제 사용하기

■ 정형화된 데이터 추출시 참조 예제를 사용하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


import uuid

from pydantic                import BaseModel
from typing                  import Optional
from pydantic                import Field
from typing                  import List
from typing                  import TypedDict
from langchain_core.messages import BaseMessage
from langchain_core.messages import HumanMessage
from langchain_core.messages import AIMessage
from langchain_core.messages import ToolMessage
from langchain_core.prompts  import ChatPromptTemplate
from langchain_core.prompts  import MessagesPlaceholder
from langchain_openai        import ChatOpenAI

class Person(BaseModel):
    """Information about a person."""

    # ^ Doc-string for the entity Person.
    # This doc-string is sent to the LLM as the description of the schema Person, and it can help to improve extraction results.

    # Note that :
    # 1. Each field is an `optional`    -- this allows the model to decline to extract it!
    # 2. Each field has a `description` -- this description is used by the LLM.
    # Having a good description can help improve extraction results.
    name             : Optional[str] = Field(..., description = "The name of the person"                 )
    hair_color       : Optional[str] = Field(..., description = "The color of the person's hair if known")
    height_in_meters : Optional[str] = Field(..., description = "Height in METERs"                       )

class Data(BaseModel):
    """Extracted data about people."""

    # Creates a model so that we can extract multiple entities.
    people : List[Person]

class Example(TypedDict):
    """A representation of an example consisting of text input and expected tool calls.

    For extraction, the tool calls are represented as instances of pydantic model.
    """

    input          : str             # This is the example text
    tool_call_list : List[BaseModel] # Instances of pydantic model that should be extracted

def getMessageList(example : Example) -> List[BaseMessage]:
    """Convert an example into a list of messages that can be fed into an LLM.

    This code is an adapter that converts our example to a list of messages that can be fed into a chat model.

    The list of messages per example corresponds to:

    1) HumanMessage : contains the content from which content should be extracted.
    2) AIMessage    : contains the extracted information from the model
    3) ToolMessage  : contains confirmation to the model that the model requested a tool correctly.

    The ToolMessage is required because some of the chat models are hyper-optimized for agents rather than for an extraction use case.
    """
    messageList : List[BaseMessage] = [HumanMessage(content = example["input"])]
    toolCallList = []
    for toolCall in example["tool_call_list"]:
        toolCallList.append(
            {
                "id"   : str(uuid.uuid4()),
                "args" : toolCall.model_dump(),
                # The name of the function right now corresponds to the name of the pydantic model
                # This is implicit in the API right now, and will be improved over time.
                "name" : toolCall.__class__.__name__
            }
        )
    messageList.append(AIMessage(content = "", tool_calls = toolCallList))
    toolOutputList = example.get("tool_outputs") or ["You have correctly called this tool."] * len(toolCallList)
    for toolOutput, toolCall in zip(toolOutputList, toolCallList):
        toolMessage = ToolMessage(content = toolOutput, tool_call_id = toolCall["id"])
        messageList.append(toolMessage)
    print(messageList)
    return messageList

exampleTupleList = [
    (
        "The ocean is vast and blue. It's more than 20,000 feet deep. There are many fish in it.",
        Data(people = []),
    ),
    (
        "Fiona traveled far from France to Spain.",
        Data(people = [Person(name = "Fiona", height_in_meters = None, hair_color = None)])
    )
]

messageList = []

for text, data in exampleTupleList:
    messageList.extend(getMessageList({"input" : text, "tool_call_list" : [data]}))

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system", "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value."),
        MessagesPlaceholder("examples"),
        ("human", "{text}")
    ]
)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini", temperature = 0)

runnableSequence1 = chatOpenAI.with_structured_output(schema = Data, method = "function_calling", include_raw = False)

runnableSequence2 = chatPromptTemplate | runnableSequence1

responseData = runnableSequence2.invoke(
    {
        "text"     : "My name is Harrison. My hair is black.",
        "examples" : messageList,
    }
)

print(responseData)

"""
people=[Person(name='Harrison', hair_color='black', height_in_meters=None)]
"""

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

import uuid

from pydantic import BaseModel

from typing import Optional

from pydantic import Field

from typing import List

from typing import TypedDict

from langchain_core.messages import BaseMessage

from langchain_core.messages import HumanMessage

from langchain_core.messages import AIMessage

from langchain_core.messages import ToolMessage

from langchain_core.prompts import ChatPromptTemplate

from langchain_core.prompts import MessagesPlaceholder

from langchain_openai import ChatOpenAI

class Person(BaseModel):

"""Information about a person."""

# ^ Doc-string for the entity Person.

# This doc-string is sent to the LLM as the description of the schema Person, and it can help to improve extraction results.

# Note that :

# 1. Each field is an `optional` -- this allows the model to decline to extract it!

# 2. Each field has a `description` -- this description is used by the LLM.

# Having a good description can help improve extraction results.

name : Optional[str] = Field(..., description = "The name of the person" )

hair_color : Optional[str] = Field(..., description = "The color of the person's hair if known")

height_in_meters : Optional[str] = Field(..., description = "Height in METERs" )

class Data(BaseModel):

"""Extracted data about people."""

# Creates a model so that we can extract multiple entities.

people : List[Person]

class Example(TypedDict):

"""A representation of an example consisting of text input and expected tool calls.

For extraction, the tool calls are represented as instances of pydantic model.

"""

input : str # This is the example text

tool_call_list : List[BaseModel] # Instances of pydantic model that should be extracted

def getMessageList(example : Example) -> List[BaseMessage]:

"""Convert an example into a list of messages that can be fed into an LLM.

This code is an adapter that converts our example to a list of messages that can be fed into a chat model.

The list of messages per example corresponds to:

1) HumanMessage : contains the content from which content should be extracted.

2) AIMessage : contains the extracted information from the model

3) ToolMessage : contains confirmation to the model that the model requested a tool correctly.

The ToolMessage is required because some of the chat models are hyper-optimized for agents rather than for an extraction use case.

"""

messageList : List[BaseMessage] = [HumanMessage(content = example["input"])]

toolCallList = []

for toolCall in example["tool_call_list"]:

toolCallList.append(

{

"id" : str(uuid.uuid4()),

"args" : toolCall.model_dump(),

# The name of the function right now corresponds to the name of the pydantic model

# This is implicit in the API right now, and will be improved over time.

"name" : toolCall.__class__.__name__

}

)

messageList.append(AIMessage(content = "", tool_calls = toolCallList))

toolOutputList = example.get("tool_outputs") or ["You have correctly called this tool."] * len(toolCallList)

for toolOutput, toolCall in zip(toolOutputList, toolCallList):

toolMessage = ToolMessage(content = toolOutput, tool_call_id = toolCall["id"])

messageList.append(toolMessage)

print(messageList)

return messageList

exampleTupleList = [

(

"The ocean is vast and blue. It's more than 20,000 feet deep. There are many fish in it.",

Data(people = []),

(

"Fiona traveled far from France to Spain.",

Data(people = [Person(name = "Fiona", height_in_meters = None, hair_color = None)])

)

]

messageList = []

for text, data in exampleTupleList:

messageList.extend(getMessageList({"input" : text, "tool_call_list" : [data]}))

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

("system", "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value."),

MessagesPlaceholder("examples"),

("human", "{text}")

]

)

chatOpenAI = ChatOpenAI(model = "gpt-4o-mini", temperature = 0)

runnableSequence1 = chatOpenAI.with_structured_output(schema = Data, method = "function_calling", include_raw = False)

runnableSequence2 = chatPromptTemplate | runnableSequence1

responseData = runnableSequence2.invoke(

{

"text" : "My name is Harrison. My hair is black.",

"examples" : messageList,

}

)

print(responseData)

"""

people=[Person(name='Harrison', hair_color='black', height_in_meters=None)]

"""

▶ requirements.txt

[PYTHON/LANGCHAIN] ChatPromptTemplate 클래스 : invoke 메소드를 사용해 ChatPromptValue 객체 만들기

■ ChatPromptTemplate 클래스의 invoke 메소드를 사용해 ChatPromptValue 객체를 만드는 방법을 보여준다. ▶ main.py


from langchain_core.prompts  import ChatPromptTemplate
from langchain_core.prompts  import MessagesPlaceholder
from langchain_core.messages import HumanMessage

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system", "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value."),
        MessagesPlaceholder("examples"),
        ("human", "{text}")
    ]
)

chatPromptValue = chatPromptTemplate.invoke({"text" : "this is some text", "examples" : [HumanMessage(content = "testing 1 2 3")]})

print(chatPromptValue)

"""
messages = [
    SystemMessage(
        content           = "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value.",
        additional_kwargs = {},
        response_metadata = {}
    ),
    HumanMessage(
        content           = 'testing 1 2 3',
        additional_kwargs = {},
        response_metadata = {}
    ),
    HumanMessage(
        content           = 'this is some text',
        additional_kwargs = {},
        response_metadata = {}
    )
]
"""

from langchain_core.prompts import ChatPromptTemplate

from langchain_core.prompts import MessagesPlaceholder

from langchain_core.messages import HumanMessage

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

MessagesPlaceholder("examples"),

("human", "{text}")

]

)

chatPromptValue = chatPromptTemplate.invoke({"text" : "this is some text", "examples" : [HumanMessage(content = "testing 1 2 3")]})

print(chatPromptValue)

"""

messages = [

SystemMessage(

content = "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value.",

additional_kwargs = {},

response_metadata = {}

HumanMessage(

content = 'testing 1 2 3',

additional_kwargs = {},

response_metadata = {}

HumanMessage(

content = 'this is some text',

additional_kwargs = {},

response_metadata = {}

)

]

"""

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==2.1.3
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==2.1.3

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain 명령을

[PYTHON/LANGCHAIN] ChatPromptTemplate 클래스 : from_messages 정적 메소드를 사용해 ChatPromptTemplate 객체 만들기

■ ChatPromptTemplate 클래스의 from_messages 정적 메소드를 사용해 ChatPromptTemplate 객체를 만드는 방법을 보여준다. ▶ main.py


from langchain_core.prompts import ChatPromptTemplate
from langchain_core.prompts import MessagesPlaceholder

chatPromptTemplate = ChatPromptTemplate.from_messages(
    [
        ("system", "You are an expert extraction algorithm. Only extract relevant information from the text. If you do not know the value of an attribute asked to extract, return null for the attribute's value."),
        MessagesPlaceholder("examples"),
        ("human", "{text}")
    ]
)

from langchain_core.prompts import ChatPromptTemplate

from langchain_core.prompts import MessagesPlaceholder

chatPromptTemplate = ChatPromptTemplate.from_messages(

[

MessagesPlaceholder("examples"),

("human", "{text}")

]

)

▶ requirements.txt


aiohappyeyeballs==2.4.4
aiohttp==3.11.9
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.6.2.post1
attrs==24.2.0
certifi==2024.8.30
charset-normalizer==3.4.0
frozenlist==1.5.0
greenlet==3.1.1
h11==0.14.0
httpcore==1.0.7
httpx==0.28.0
idna==3.10
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.3.9
langchain-core==0.3.21
langchain-text-splitters==0.3.2
langsmith==0.1.147
multidict==6.1.0
numpy==2.1.3
orjson==3.10.12
packaging==24.2
propcache==0.2.1
pydantic==2.10.3
pydantic_core==2.27.1
PyYAML==6.0.2
requests==2.32.3
requests-toolbelt==1.0.0
sniffio==1.3.1
SQLAlchemy==2.0.36
tenacity==9.0.0
typing_extensions==4.12.2
urllib3==2.2.3
yarl==1.18.3

aiohappyeyeballs==2.4.4

aiohttp==3.11.9

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.6.2.post1

attrs==24.2.0

certifi==2024.8.30

charset-normalizer==3.4.0

frozenlist==1.5.0

greenlet==3.1.1

h11==0.14.0

httpcore==1.0.7

httpx==0.28.0

idna==3.10

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.3.9

langchain-core==0.3.21

langchain-text-splitters==0.3.2

langsmith==0.1.147

multidict==6.1.0

numpy==2.1.3

orjson==3.10.12

packaging==24.2

propcache==0.2.1

pydantic==2.10.3

pydantic_core==2.27.1

PyYAML==6.0.2

requests==2.32.3

requests-toolbelt==1.0.0

sniffio==1.3.1

SQLAlchemy==2.0.36

tenacity==9.0.0

typing_extensions==4.12.2

urllib3==2.2.3

yarl==1.18.3

※ pip install langchain

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : invoke 메소드에서 config 인자를 사용해 벡터 저장소 검색하기

■ RunnableSequence 클래스의 invoke 메소드에서 config 인자를 사용해 벡터 저장소를 검색하는 방법을 보여준다. ※ PINECONE_API_KEY 환경 변수 값은 .env 파일에 정의한다. ※

[PYTHON/LANGCHAIN] RunnableConfigurableFields 클래스 : invoke 메소드에서 config 인자를 사용해 벡터 유사도 검색하기

■ RunnableConfigurableFields 클래스의 invoke 메소드에서 config 인자를 사용해 벡터 유사도를 검색하는 방법을 보여준다. ※ PINECONE_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶