VECTOR STORE Archives - 4 중 3 번째 페이지

[PYTHON/LANGCHAIN] FAISS 클래스 : as_retriever 메소드를 사용해 벡터 스토어 검색기 구하기

■ FAISS 클래스의 as_retriever 메소드를 사용해 벡터 스토어 검색기를 구하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ 예제

[PYTHON/LANGCHAIN] Chroma 클래스 : similarity_search_by_vector 메소드를 사용해 검색 문자열 벡터 리스트로 검색 결과 문서 리스트 구하기 2

■ Chroma 클래스의 similarity_search_by_vector 메소드를 사용해 검색 문자열의 벡터 리스트로 검색 결과 문서 리스트를 구하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은

[PYTHON/LANGCHAIN] FAISS 클래스 : similarity_search 메소드를 사용해 문자열 쿼리 유사성 기반으로 문서 리스트 구하기

■ FAISS 클래스의 similarity_search 메소드를 사용해 문자열 쿼리 유사성 기반으로 문서 리스트를 구하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에

[PYTHON/LANGCHAIN] from_documents 함수 : 메모리 내 벡터 저장소를 위한 FAISS 객체 만들기

■ from_documents 함수를 사용해 메모리 내 벡터 저장소를 위한 FAISS 객체를 만드는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다.

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : get_prompts 메소드를 사용해 체인에서 사용하는 프롬프트 템플리트 리스트 구하기

■ RunnableSequence 클래스의 get_prompts 메소드를 사용해 체인에서 사용하는 프롬프트 템플리트 리스트를 구하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다.

[PYTHON/LANGCHAIN] Graph 클래스 : print_ascii 메소드를 사용해 그래프 인쇄하기

■ Graph 클래스의 print_ascii 메소드를 사용해 그래프를 인쇄하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from dotenv                           import load_dotenv
from langchain_community.vectorstores import FAISS
from langchain_openai                 import OpenAIEmbeddings
from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import ChatOpenAI
from langchain_core.runnables         import RunnablePassthrough
from langchain_core.output_parsers    import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :
{context}

Question : {question}
"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

runnableSequence = {"context" : vectorStoreRetriever, "question" : RunnablePassthrough()} | chatPromptTemplate | chatOpenAI | StrOutputParser()

graph = runnableSequence.get_graph()

graph.print_ascii()

"""
           +---------------------------------+
           | Parallel<context,question>Input |
           +---------------------------------+
                    **               **
                 ***                   ***
               **                         **
+----------------------+              +-------------+
| VectorStoreRetriever |              | Passthrough |
+----------------------+              +-------------+
                    **               **
                      ***         ***
                         **     **
           +----------------------------------+
           | Parallel<context,question>Output |
           +----------------------------------+
                             *
                             *
                             *
                  +--------------------+
                  | ChatPromptTemplate |
                  +--------------------+
                             *
                             *
                             *
                      +------------+
                      | ChatOpenAI |
                      +------------+
                             *
                             *
                             *
                   +-----------------+
                   | StrOutputParser |
                   +-----------------+
                             *
                             *
                             *
                +-----------------------+
                | StrOutputParserOutput |
                +-----------------------+
"""

from dotenv import load_dotenv

from langchain_community.vectorstores import FAISS

from langchain_openai import OpenAIEmbeddings

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

from langchain_core.runnables import RunnablePassthrough

from langchain_core.output_parsers import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

runnableSequence = {"context" : vectorStoreRetriever, "question" : RunnablePassthrough()} | chatPromptTemplate | chatOpenAI | StrOutputParser()

graph = runnableSequence.get_graph()

graph.print_ascii()

"""

+---------------------------------+

| Parallel<context,question>Input |

+---------------------------------+

** **

*** ***

** **

+----------------------+ +-------------+

| VectorStoreRetriever | | Passthrough |

+----------------------+ +-------------+

** **

*** ***

** **

+----------------------------------+

| Parallel<context,question>Output |

+----------------------------------+

+--------------------+

| ChatPromptTemplate |

+--------------------+

+------------+

| ChatOpenAI |

+------------+

+-----------------+

| StrOutputParser |

+-----------------+

+-----------------------+

| StrOutputParserOutput |

+-----------------------+

"""

▶

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : get_graph 메소드를 사용해 Graph 객체 구하기

■ RunnableSequence 클래스의 get_graph 메소드를 사용해 Graph 객체를 구하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from dotenv                           import load_dotenv
from langchain_community.vectorstores import FAISS
from langchain_openai                 import OpenAIEmbeddings
from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import ChatOpenAI
from langchain_core.runnables         import RunnablePassthrough
from langchain_core.output_parsers    import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :
{context}

Question : {question}
"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

runnableSequence = {"context" : vectorStoreRetriever, "question" : RunnablePassthrough()} | chatPromptTemplate | chatOpenAI | StrOutputParser()

graph = runnableSequence.get_graph()

from dotenv import load_dotenv

from langchain_community.vectorstores import FAISS

from langchain_openai import OpenAIEmbeddings

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

from langchain_core.runnables import RunnablePassthrough

from langchain_core.output_parsers import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

runnableSequence = {"context" : vectorStoreRetriever, "question" : RunnablePassthrough()} | chatPromptTemplate | chatOpenAI | StrOutputParser()

graph = runnableSequence.get_graph()

[PYTHON/LANGCHAIN] RunnablePassthrough 클래스 : 검색기와 함께 사용자 질문을 채팅 프롬프트 템플리트에 전달하기

■ RunnablePassthrough 클래스를 사용해 검색기와 함께 사용자 질문을 채팅 프롬프트 템플리트에 전달하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다.

[PYTHON/LANGCHAIN] RunnableParallel 클래스 : itemgetter 객체를 사용해 딕셔너리에서 데이터 추출하기

■ RunnableParallel 클래스에서 itemgetter 객체를 사용해 딕셔너리에서 데이터를 추출하는 방법을 보여준다. ※ OPENAI_API_KEY 환경 변수 값은 .env 파일에 정의한다. ▶ main.py


from operator import itemgetter

from dotenv                           import load_dotenv
from langchain_community.vectorstores import FAISS
from langchain_openai                 import OpenAIEmbeddings
from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :
{context}

Question : {question}

Answer in the following language : {language}
"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

chain = (
    # { ... } 부분은 RunnableParallel 클래스로 자동 변환된다.
    {"context" : itemgetter("question") | vectorStoreRetriever, "question" : itemgetter("question"), "language" : itemgetter("language")}
    | chatPromptTemplate
    | chatOpenAI
    | StrOutputParser()
)

responseString = chain.invoke({"question" : "where did harrison work", "language" : "korean"})

print(responseString)

"""
헤리슨은 켄쇼에서 일했습니다.
"""

from operator import itemgetter

from dotenv import load_dotenv

from langchain_community.vectorstores import FAISS

from langchain_openai import OpenAIEmbeddings

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

load_dotenv()

faiss = FAISS.from_texts(["harrison worked at kensho"], embedding = OpenAIEmbeddings())

vectorStoreRetriever = faiss.as_retriever()

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

Answer in the following language : {language}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

chatOpenAI = ChatOpenAI()

chain = (

# { ... } 부분은 RunnableParallel 클래스로 자동 변환된다.

{"context" : itemgetter("question") | vectorStoreRetriever, "question" : itemgetter("question"), "language" : itemgetter("language")}

| chatPromptTemplate

| chatOpenAI

| StrOutputParser()

)

responseString = chain.invoke({"question" : "where did harrison work", "language" : "korean"})

print(responseString)

"""

헤리슨은 켄쇼에서 일했습니다.

"""

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드에서 마지막 비스트리밍 단계 이후 비동기 스트리밍 이벤트 수신하기

■ RunnableSequence 클래스의 astream_events 메소드에서 마지막 비스트리밍 단계 이후 비동기 스트리밍 이벤트를 수신하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_openai              import ChatOpenAI
from langchain_core.output_parsers import JsonOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    jsonOutputParser = JsonOutputParser()

    def getCountryNameList(inputDictionary):
        """입력 스트림에서 작동하지 않고 스트리밍을 중단하는 함수이다."""
        if not isinstance(inputDictionary, dict):
            return ""
        if "countries" not in inputDictionary:
            return ""
        countryList = inputDictionary["countries"]
        if not isinstance(countryList, list):
            return ""
        countryNameList = [country.get("name") for country in countryList if isinstance(country, dict)]
        return countryNameList

    runnableSequence = chatOpenAI | jsonOutputParser | getCountryNameList

    requestString = """output a list of the countries france, spain and japan and their populations in JSON format. \
    Use a dict with an outer key of "countries" which contains a list of countries. \
    Each country should have the key `name` and `population`"""

    async for chunkList in runnableSequence.astream(requestString):
        print(chunkList, flush = True)

    eventCount = 0

    async for eventDictionary in runnableSequence.astream_events(requestString, version = "v2"):
        kind = eventDictionary["event"]

        if kind == "on_chat_model_stream":
            content = eventDictionary["data"]["chunk"].content
            print(f"Chat model chunk : {content}", flush = True)
        if kind == "on_parser_stream":
            text = eventDictionary["data"]["chunk"]
            print(f"Parser chunk     : {text}", flush = True)
        eventCount += 1
        if eventCount > 30:
            print("...")
            break

asyncio.run(main())

"""
Chat model chunk :
Chat model chunk : {

Parser chunk     : {}
Chat model chunk :
Chat model chunk :  "
Chat model chunk : countries
Chat model chunk : ":
Chat model chunk :  [

Parser chunk     : {'countries': []}
Chat model chunk :
Chat model chunk :  {

Parser chunk     : {'countries': [{}]}
Chat model chunk :
Chat model chunk :  "
Chat model chunk : name
Chat model chunk : ":
Chat model chunk :  "
Parser chunk     : {'countries': [{'name': ''}]}
Chat model chunk : France
Parser chunk     : {'countries': [{'name': 'France'}]}
Chat model chunk : ",

Chat model chunk :
Chat model chunk :  "
Chat model chunk : population
Chat model chunk : ":
Chat model chunk :
Chat model chunk : 652
...
"""

import asyncio

import os

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import JsonOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

jsonOutputParser = JsonOutputParser()

def getCountryNameList(inputDictionary):

"""입력 스트림에서 작동하지 않고 스트리밍을 중단하는 함수이다."""

if not isinstance(inputDictionary, dict):

return ""

if "countries" not in inputDictionary:

return ""

countryList = inputDictionary["countries"]

if not isinstance(countryList, list):

return ""

countryNameList = [country.get("name") for country in countryList if isinstance(country, dict)]

return countryNameList

runnableSequence = chatOpenAI | jsonOutputParser | getCountryNameList

requestString = """output a list of the countries france, spain and japan and their populations in JSON format. \

Use a dict with an outer key of "countries" which contains a list of countries. \

Each country should have the key `name` and `population`"""

async for chunkList in runnableSequence.astream(requestString):

print(chunkList, flush = True)

eventCount = 0

async for eventDictionary in runnableSequence.astream_events(requestString, version = "v2"):

kind = eventDictionary["event"]

if kind == "on_chat_model_stream":

content = eventDictionary["data"]["chunk"].content

print(f"Chat model chunk : {content}", flush = True)

if kind == "on_parser_stream":

text = eventDictionary["data"]["chunk"]

print(f"Parser chunk : {text}", flush = True)

eventCount += 1

if eventCount > 30:

print("...")

break

asyncio.run(main())

"""

Chat model chunk :

Chat model chunk : {

Parser chunk : {}

Chat model chunk :

Chat model chunk : "

Chat model chunk : countries

Chat model chunk : ":

Chat model chunk : [

Parser chunk : {'countries': []}

Chat model chunk :

Chat model chunk : {

Parser chunk : {'countries': [{}]}

Chat model chunk :

Chat model chunk : "

Chat model chunk : name

Chat model chunk : ":

Chat model chunk : "

Parser chunk : {'countries': [{'name': ''}]}

Chat model chunk : France

Parser chunk : {'countries': [{'name': 'France'}]}

Chat model chunk : ",

Chat model chunk :

Chat model chunk : "

Chat model chunk : population

Chat model chunk : ":

Chat model chunk :

Chat model chunk : 652

...

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

※

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드에서 include_tags 인자를 사용해 이벤트 필터링하기

■ RunnableSequence 클래스의 astream_events 메소드에서 include_tags 인자를 사용해 이벤트를 필터링하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.runnables         import RunnablePassthrough
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatPromptTemplateString = """Answer the question based only on the following context :
    {context}

    Question : {question}
    """

    chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

    faiss = FAISS.from_texts(
        ["harrison worked at kensho", "harrison likes spicy food"],
        embedding = OpenAIEmbeddings(),
    )

    vectorStoreRetriever = faiss.as_retriever()

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    runnableSequence = (
        {
            "context"  : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),
            "question" : RunnablePassthrough()
        }
        | chatPromptTemplate
        | (chatOpenAI.with_config(run_name = "chatOpenAI") | StrOutputParser().with_config(run_name = "strOutputParser")).with_config({"tags" : ["chatOpenAI_strOutputParser"]})
    )

    eventCount = 0

    async for eventDictionary in runnableSequence.astream_events(
        "Where did harrison work? Write 3 made up sentences about this place.",
        include_tags = ["chatOpenAI_strOutputParser"],
        version = "v2"):
        print(eventDictionary)
        eventCount += 1
        if eventCount > 10:
            print("...")
            break

asyncio.run(main())

"""
{'event': 'on_chain_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'metadata': {}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}
{'event': 'on_chat_model_start', 'data': {'input': {'messages': [[HumanMessage(content="Answer the question based only on the following context :\n    [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]\n\n    Question : Where did harrison work? Write 3 made up sentences about this place.\n    ")]]}}, 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_parser_start', 'data': {}, 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'metadata': {}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_chain_stream', 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='H', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_chain_stream', 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='arrison', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'arrison'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}
...
"""

import asyncio

import os

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import OpenAIEmbeddings

from langchain_community.vectorstores import FAISS

from langchain_core.runnables import RunnablePassthrough

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

faiss = FAISS.from_texts(

["harrison worked at kensho", "harrison likes spicy food"],

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = faiss.as_retriever()

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

runnableSequence = (

{

"context" : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),

"question" : RunnablePassthrough()

}

| chatPromptTemplate

| (chatOpenAI.with_config(run_name = "chatOpenAI") | StrOutputParser().with_config(run_name = "strOutputParser")).with_config({"tags" : ["chatOpenAI_strOutputParser"]})

)

eventCount = 0

async for eventDictionary in runnableSequence.astream_events(

"Where did harrison work? Write 3 made up sentences about this place.",

include_tags = ["chatOpenAI_strOutputParser"],

version = "v2"):

print(eventDictionary)

eventCount += 1

if eventCount > 10:

print("...")

break

asyncio.run(main())

"""

{'event': 'on_chain_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'metadata': {}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}

{'event': 'on_chat_model_start', 'data': {'input': {'messages': [[HumanMessage(content="Answer the question based only on the following context :\n [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]\n\n Question : Where did harrison work? Write 3 made up sentences about this place.\n ")]]}}, 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_parser_start', 'data': {}, 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'metadata': {}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_chain_stream', 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='H', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_chain_stream', 'run_id': '8933aa37-93c0-4c63-ae77-1c568c2f5a7a', 'name': 'RunnableSequence', 'tags': ['seq:step:3', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='arrison', id='run-35559cbb-163c-46cf-b278-fbda9d987add')}, 'run_id': '35559cbb-163c-46cf-b278-fbda9d987add', 'name': 'chatOpenAI', 'tags': ['seq:step:1', 'chatOpenAI_strOutputParser'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

{'event': 'on_parser_stream', 'run_id': '7c5ca72a-9681-49a0-b33d-23dc26a1e245', 'name': 'strOutputParser', 'tags': ['seq:step:2', 'chatOpenAI_strOutputParser'], 'metadata': {}, 'data': {'chunk': 'arrison'}, 'parent_ids': ['bf69804f-1597-4a2a-9ad8-faaff7c767c2', '8933aa37-93c0-4c63-ae77-1c568c2f5a7a']}

...

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

※ pip install langchain

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드에서 include_types 인자를 사용해 이벤트 필터링하기

■ RunnableSequence 클래스의 astream_events 메소드에서 include_types 인자를 사용해 이벤트를 필터링하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.runnables         import RunnablePassthrough
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatPromptTemplateString = """Answer the question based only on the following context :
    {context}

    Question : {question}
    """

    chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

    faiss = FAISS.from_texts(
        ["harrison worked at kensho", "harrison likes spicy food"],
        embedding = OpenAIEmbeddings(),
    )

    vectorStoreRetriever = faiss.as_retriever()

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    runnableSequence = (
        {
            "context"  : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),
            "question" : RunnablePassthrough()
        }
        | chatPromptTemplate
        | chatOpenAI.with_config(run_name = "chatOpenAI")
        | StrOutputParser().with_config(run_name = "strOutputParser")
    )

    eventCount = 0

    async for eventDictionary in runnableSequence.astream_events(
        "Where did harrison work? Write 3 made up sentences about this place.",
        include_types = ["chat_model"],
        version = "v2"):
        print(eventDictionary)
        eventCount += 1
        if eventCount > 10:
            print("...")
            break

asyncio.run(main())

"""
{'event': 'on_chat_model_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='H', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='arrison', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' worked', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' at', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' Kens', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='ho', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=',', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' a', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' bustling', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}
...
"""

import asyncio

import os

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import OpenAIEmbeddings

from langchain_community.vectorstores import FAISS

from langchain_core.runnables import RunnablePassthrough

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

faiss = FAISS.from_texts(

["harrison worked at kensho", "harrison likes spicy food"],

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = faiss.as_retriever()

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

runnableSequence = (

{

"context" : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),

"question" : RunnablePassthrough()

}

| chatPromptTemplate

| chatOpenAI.with_config(run_name = "chatOpenAI")

| StrOutputParser().with_config(run_name = "strOutputParser")

)

eventCount = 0

async for eventDictionary in runnableSequence.astream_events(

"Where did harrison work? Write 3 made up sentences about this place.",

include_types = ["chat_model"],

version = "v2"):

print(eventDictionary)

eventCount += 1

if eventCount > 10:

print("...")

break

asyncio.run(main())

"""

{'event': 'on_chat_model_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='H', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='arrison', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' worked', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' at', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' Kens', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='ho', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=',', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' a', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content=' bustling', id='run-ba5d2b4e-a267-407a-b028-c42ccbcb9d63')}, 'run_id': 'ba5d2b4e-a267-407a-b028-c42ccbcb9d63', 'name': 'chatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['be3b9431-47a9-40fd-ac10-4c6dcc29a0d4']}

...

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

※ pip install langchain

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드에서 include_names 인자를 사용해 이벤트 필터링하기

■ RunnableSequence 클래스의 astream_events 메소드에서 include_names 인자를 사용해 이벤트를 필터링하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.runnables         import RunnablePassthrough
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatPromptTemplateString = """Answer the question based only on the following context :
    {context}

    Question : {question}
    """

    chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

    faiss = FAISS.from_texts(
        ["harrison worked at kensho", "harrison likes spicy food"],
        embedding = OpenAIEmbeddings(),
    )

    vectorStoreRetriever = faiss.as_retriever()

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    runnableSequence = (
        {
            "context"  : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),
            "question" : RunnablePassthrough()
        }
        | chatPromptTemplate
        | chatOpenAI.with_config(run_name = "chatOpenAI")
        | StrOutputParser().with_config(run_name = "strOutputParser")
    )

    eventCount = 0

    async for eventDictionary in runnableSequence.astream_events(
        "Where did harrison work? Write 3 made up sentences about this place.",
        include_names = ["strOutputParser"],
        version = "v2"):
        print(eventDictionary)
        eventCount += 1
        if eventCount > 10:
            print("...")
            break

asyncio.run(main())

"""
{'event': 'on_parser_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'metadata': {}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': 'arrison'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' worked'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' at'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' a'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' popular'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' restaurant'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' called'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}
...
"""

import asyncio

import os

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import OpenAIEmbeddings

from langchain_community.vectorstores import FAISS

from langchain_core.runnables import RunnablePassthrough

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

faiss = FAISS.from_texts(

["harrison worked at kensho", "harrison likes spicy food"],

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = faiss.as_retriever()

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

runnableSequence = (

{

"context" : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),

"question" : RunnablePassthrough()

}

| chatPromptTemplate

| chatOpenAI.with_config(run_name = "chatOpenAI")

| StrOutputParser().with_config(run_name = "strOutputParser")

)

eventCount = 0

async for eventDictionary in runnableSequence.astream_events(

"Where did harrison work? Write 3 made up sentences about this place.",

include_names = ["strOutputParser"],

version = "v2"):

print(eventDictionary)

eventCount += 1

if eventCount > 10:

print("...")

break

asyncio.run(main())

"""

{'event': 'on_parser_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'metadata': {}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': 'H'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': 'arrison'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' worked'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' at'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' a'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' popular'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' restaurant'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

{'event': 'on_parser_stream', 'run_id': '99be2832-6fb2-4482-af73-93bb583db3df', 'name': 'strOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ' called'}, 'parent_ids': ['735f3142-4099-43ca-bd0e-c141bd5ab020']}

...

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

※ pip install langchain

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드를 사용해 비동기 스트리밍 이벤트 수신시 이벤트 종류에 따라 처리하기

■ RunnableSequence 클래스의 astream_events 메소드를 사용해 비동기 스트리밍 이벤트 수신시 이벤트 종류에 따라 처리하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.runnables         import RunnablePassthrough
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatPromptTemplateString = """Answer the question based only on the following context :
    {context}

    Question : {question}
    """

    chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

    faiss = FAISS.from_texts(
        ["harrison worked at kensho", "harrison likes spicy food"],
        embedding = OpenAIEmbeddings(),
    )

    vectorStoreRetriever = faiss.as_retriever()

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    runnableSequence = (
        {
            "context"  : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),
            "question" : RunnablePassthrough()
        }
        | chatPromptTemplate
        | chatOpenAI
        | StrOutputParser()
    )

    eventCount = 0

    async for eventDictionary in runnableSequence.astream_events("Where did harrison work? Write 3 made up sentences about this place.", version = "v2"):
        kind = eventDictionary["event"]
        if kind == "on_chat_model_stream":
            content = eventDictionary["data"]["chunk"].content
            print(f"Chat model chunk : {content}", flush = True)
        if kind == "on_parser_stream":
            text = eventDictionary["data"]["chunk"]
            print(f"Parser chunk     : {text}", flush = True)
        eventCount += 1
        if eventCount > 30:
            print("...")
            break

asyncio.run(main())

"""
Chat model chunk :
Parser chunk     :
Chat model chunk : H
Parser chunk     : H
Chat model chunk : arrison
Parser chunk     : arrison
Chat model chunk :  worked
Parser chunk     :  worked
Chat model chunk :  at
Parser chunk     :  at
Chat model chunk :  a
Parser chunk     :  a
...
"""

import asyncio

import os

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import OpenAIEmbeddings

from langchain_community.vectorstores import FAISS

from langchain_core.runnables import RunnablePassthrough

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

faiss = FAISS.from_texts(

["harrison worked at kensho", "harrison likes spicy food"],

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = faiss.as_retriever()

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

runnableSequence = (

{

"context" : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),

"question" : RunnablePassthrough()

}

| chatPromptTemplate

| chatOpenAI

| StrOutputParser()

)

eventCount = 0

async for eventDictionary in runnableSequence.astream_events("Where did harrison work? Write 3 made up sentences about this place.", version = "v2"):

kind = eventDictionary["event"]

if kind == "on_chat_model_stream":

content = eventDictionary["data"]["chunk"].content

print(f"Chat model chunk : {content}", flush = True)

if kind == "on_parser_stream":

text = eventDictionary["data"]["chunk"]

print(f"Parser chunk : {text}", flush = True)

eventCount += 1

if eventCount > 30:

print("...")

break

asyncio.run(main())

"""

Chat model chunk :

Parser chunk :

Chat model chunk : H

Parser chunk : H

Chat model chunk : arrison

Parser chunk : arrison

Chat model chunk : worked

Parser chunk : worked

Chat model chunk : at

Parser chunk : at

Chat model chunk : a

Parser chunk : a

...

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

[PYTHON/LANGCHAIN] RunnableSequence 클래스 : astream_events 메소드를 사용해 비동기 스트리밍 이벤트 수신하기

■ RunnableSequence 클래스의 astream_events 메소드를 사용해 비동기 스트리밍 이벤트를 수신하는 방법을 보여준다. ▶ main.py


import asyncio
import os

from langchain_core.prompts           import ChatPromptTemplate
from langchain_openai                 import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.runnables         import RunnablePassthrough
from langchain_openai                 import ChatOpenAI
from langchain_core.output_parsers    import StrOutputParser

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    chatPromptTemplateString = """Answer the question based only on the following context :
    {context}

    Question : {question}
    """

    chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

    faiss = FAISS.from_texts(
        ["harrison worked at kensho", "harrison likes spicy food"],
        embedding = OpenAIEmbeddings(),
    )

    vectorStoreRetriever = faiss.as_retriever()

    chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

    runnableSequence = (
        {
            "context"  : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),
            "question" : RunnablePassthrough()
        }
        | chatPromptTemplate
        | chatOpenAI
        | StrOutputParser()
    )

    async for eventDictionary in runnableSequence.astream_events("Where did harrison work? Write 3 made up sentences about this place.", version = "v2"):
        print(eventDictionary)

asyncio.run(main())

"""
{'event': 'on_chain_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'RunnableSequence', 'tags': [], 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'metadata': {}, 'parent_ids': []}
{'event': 'on_chain_start', 'data': {}, 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_retriever_start', 'data': {'input': {'query': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'name': 'Docs', 'tags': ['map:key:context', 'FAISS', 'OpenAIEmbeddings'], 'run_id': '892860bc-24a2-41d2-adc8-1572931c2c8c', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}
{'event': 'on_chain_start', 'data': {}, 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}
{'event': 'on_chain_stream', 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'metadata': {}, 'data': {'chunk': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}
{'event': 'on_chain_stream', 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'metadata': {}, 'data': {'chunk': {'question': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_chain_end', 'data': {'output': 'Where did harrison work? Write 3 made up sentences about this place.', 'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}
{'event': 'on_retriever_end', 'data': {'output': [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')], 'input': {'query': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'run_id': '892860bc-24a2-41d2-adc8-1572931c2c8c', 'name': 'Docs', 'tags': ['map:key:context', 'FAISS', 'OpenAIEmbeddings'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}
{'event': 'on_chain_stream', 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'metadata': {}, 'data': {'chunk': {'context': [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]}}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

...

{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': ' customers'}, 'parent_ids': []}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='.', id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_parser_stream', 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': '.'}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': '.'}, 'parent_ids': []}
{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_parser_stream', 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': []}
{'event': 'on_chat_model_end', 'data': {'output': AIMessageChunk(content='Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11'), 'input': {'messages': [[HumanMessage(content="Answer the question based only on the following context :\n    [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]\n\n    Question : Where did harrison work? Write 3 made up sentences about this place.\n    ")]]}}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_parser_end', 'data': {'output': 'Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', 'input': AIMessageChunk(content='Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}
{'event': 'on_chain_end', 'data': {'output': 'Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.'}, 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'parent_ids': []}
"""

import asyncio

import os

from langchain_core.prompts import ChatPromptTemplate

from langchain_openai import OpenAIEmbeddings

from langchain_community.vectorstores import FAISS

from langchain_core.runnables import RunnablePassthrough

from langchain_openai import ChatOpenAI

from langchain_core.output_parsers import StrOutputParser

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

chatPromptTemplateString = """Answer the question based only on the following context :

{context}

Question : {question}

"""

chatPromptTemplate = ChatPromptTemplate.from_template(chatPromptTemplateString)

faiss = FAISS.from_texts(

["harrison worked at kensho", "harrison likes spicy food"],

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = faiss.as_retriever()

chatOpenAI = ChatOpenAI(model = "gpt-3.5-turbo-0125")

runnableSequence = (

{

"context" : vectorStoreRetriever.with_config(run_name = "vectorStoreRetriever"),

"question" : RunnablePassthrough()

}

| chatPromptTemplate

| chatOpenAI

| StrOutputParser()

)

async for eventDictionary in runnableSequence.astream_events("Where did harrison work? Write 3 made up sentences about this place.", version = "v2"):

print(eventDictionary)

asyncio.run(main())

"""

{'event': 'on_chain_start', 'data': {'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'name': 'RunnableSequence', 'tags': [], 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'metadata': {}, 'parent_ids': []}

{'event': 'on_chain_start', 'data': {}, 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_retriever_start', 'data': {'input': {'query': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'name': 'Docs', 'tags': ['map:key:context', 'FAISS', 'OpenAIEmbeddings'], 'run_id': '892860bc-24a2-41d2-adc8-1572931c2c8c', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}

{'event': 'on_chain_start', 'data': {}, 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}

{'event': 'on_chain_stream', 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'metadata': {}, 'data': {'chunk': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}

{'event': 'on_chain_stream', 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'metadata': {}, 'data': {'chunk': {'question': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_chain_end', 'data': {'output': 'Where did harrison work? Write 3 made up sentences about this place.', 'input': 'Where did harrison work? Write 3 made up sentences about this place.'}, 'run_id': '5343e2d7-1d30-44c8-b599-bf3a11170ffc', 'name': 'RunnablePassthrough', 'tags': ['map:key:question'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}

{'event': 'on_retriever_end', 'data': {'output': [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')], 'input': {'query': 'Where did harrison work? Write 3 made up sentences about this place.'}}, 'run_id': '892860bc-24a2-41d2-adc8-1572931c2c8c', 'name': 'Docs', 'tags': ['map:key:context', 'FAISS', 'OpenAIEmbeddings'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611', '98a359fb-0a7b-455e-901a-819e5db32dba']}

{'event': 'on_chain_stream', 'run_id': '98a359fb-0a7b-455e-901a-819e5db32dba', 'name': 'RunnableParallel<context,question>', 'tags': ['seq:step:1'], 'metadata': {}, 'data': {'chunk': {'context': [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]}}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

...

{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': ' customers'}, 'parent_ids': []}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='.', id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_parser_stream', 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': '.'}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': '.'}, 'parent_ids': []}

{'event': 'on_chat_model_stream', 'data': {'chunk': AIMessageChunk(content='', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_parser_stream', 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_chain_stream', 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'data': {'chunk': ''}, 'parent_ids': []}

{'event': 'on_chat_model_end', 'data': {'output': AIMessageChunk(content='Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11'), 'input': {'messages': [[HumanMessage(content="Answer the question based only on the following context :\n [Document(page_content='harrison worked at kensho'), Document(page_content='harrison likes spicy food')]\n\n Question : Where did harrison work? Write 3 made up sentences about this place.\n ")]]}}, 'run_id': '8fecc191-56e5-4860-b876-6da901a96e11', 'name': 'ChatOpenAI', 'tags': ['seq:step:3'], 'metadata': {'ls_provider': 'openai', 'ls_model_name': 'gpt-3.5-turbo-0125', 'ls_model_type': 'chat', 'ls_temperature': 0.7}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_parser_end', 'data': {'output': 'Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', 'input': AIMessageChunk(content='Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.', response_metadata={'finish_reason': 'stop'}, id='run-8fecc191-56e5-4860-b876-6da901a96e11')}, 'run_id': '6959f904-9b48-42d4-8859-d9c1ae0c0b2e', 'name': 'StrOutputParser', 'tags': ['seq:step:4'], 'metadata': {}, 'parent_ids': ['d897f470-051b-41be-953a-832c5cc06611']}

{'event': 'on_chain_end', 'data': {'output': 'Harrison worked at Kensho, a trendy restaurant known for its fusion cuisine. The atmosphere at Kensho is lively and welcoming, with a modern decor that attracts a hip crowd. The menu at Kensho features a variety of dishes from around the world, all prepared with a spicy twist to cater to the adventurous palate of its customers.'}, 'run_id': 'd897f470-051b-41be-953a-832c5cc06611', 'name': 'RunnableSequence', 'tags': [], 'metadata': {}, 'parent_ids': []}

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.5
langchain-community==0.2.5
langchain-core==0.2.8
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.79
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.34.0
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.4.1
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.2
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.5

langchain-community==0.2.5

langchain-core==0.2.8

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.79

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.34.0

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.4.1

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.2

yarl==1.9.4

※ pip install langchain

[PYTHON/LANGCHAIN] VectorStoreRetriever 클래스 : 사용자 질문에 따라 벡터 저장소에서 문서 자동 검색하기

■ VectorStoreRetriever 클래스 : 사용자 질문에 따라 벡터 저장소에서 문서 자동 검색하기 ▶ main.py


from langchain_community.document_loaders import WebBaseLoader
from langchain_text_splitters             import RecursiveCharacterTextSplitter
from langchain_huggingface                import HuggingFaceEmbeddings
from langchain_chroma                     import Chroma
from langchain_community.llms             import LlamaCpp
from langchain                            import hub
from langchain_core.output_parsers        import StrOutputParser
from langchain_core.runnables             import RunnablePassthrough

# 웹 문서를 로드한다.
webBaseLoader = WebBaseLoader("https://lilianweng.github.io/posts/2023-06-23-agent/")

documentList = webBaseLoader.load()

# 문서를 분할한다.
recursiveCharacterTextSplitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)

splitDocumentList = recursiveCharacterTextSplitter.split_documents(documentList)

# 임베딩 생성 및 유사도 검색을 위한 벡터 저장소를 설정한다.
chroma = Chroma.from_documents(documents = splitDocumentList, embedding = HuggingFaceEmbeddings())

# 벡터 저장소 검색기를 설정한다.
vectorStoreRetriever = chroma.as_retriever()

# LLM 모델을 설정한다.
gpuLayerCount = 1
batchSize     = 512 # 1에서 n_ctx 사이이어야 한다.

llamaCpp = LlamaCpp(
    model_path   = "./llama-2-13b-chat.Q4_0.gguf", # llama-2-13b-chat.Q4_0.gguf 파일 경로를 설정한다.
    n_gpu_layers = gpuLayerCount,
    n_batch      = batchSize,
    n_ctx        = 2048,
    f16_kv       = True, # 반드시 True로 설정해야 한다. 그렇지 않으면 몇 번의 호출 후에 문제가 발생하게 된다.
    verbose      = False
)

# 프롬프트 템플리트를 설정한다.
promptTemplate = hub.pull("rlm/rag-prompt")

# 문서 페이지 컨텐츠를 병합하는 함수를 정의한다.
def mergeDocumentPageContent(documentList):
    return "\n\n".join(doc.page_content for doc in documentList)

# 실행 체인을 설정한다.
runnableSequence = (
    {"context": vectorStoreRetriever | mergeDocumentPageContent, "question" : RunnablePassthrough()}
    | promptTemplate
    | llamaCpp
    | StrOutputParser()
)

# 질의 응답을 실행한다.
resultString = runnableSequence.invoke("What are the approaches to Task Decomposition?")

print(resultString)

"""
 The approaches to task decomposition include breaking down large tasks into smaller subgoals, using task-specific instructions, and using human inputs.
"""

from langchain_community.document_loaders import WebBaseLoader

from langchain_text_splitters import RecursiveCharacterTextSplitter

from langchain_huggingface import HuggingFaceEmbeddings

from langchain_chroma import Chroma

from langchain_community.llms import LlamaCpp

from langchain import hub

from langchain_core.output_parsers import StrOutputParser

from langchain_core.runnables import RunnablePassthrough

# 웹 문서를 로드한다.

webBaseLoader = WebBaseLoader("https://lilianweng.github.io/posts/2023-06-23-agent/")

documentList = webBaseLoader.load()

# 문서를 분할한다.

recursiveCharacterTextSplitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)

splitDocumentList = recursiveCharacterTextSplitter.split_documents(documentList)

# 임베딩 생성 및 유사도 검색을 위한 벡터 저장소를 설정한다.

chroma = Chroma.from_documents(documents = splitDocumentList, embedding = HuggingFaceEmbeddings())

# 벡터 저장소 검색기를 설정한다.

vectorStoreRetriever = chroma.as_retriever()

# LLM 모델을 설정한다.

gpuLayerCount = 1

batchSize = 512 # 1에서 n_ctx 사이이어야 한다.

llamaCpp = LlamaCpp(

model_path = "./llama-2-13b-chat.Q4_0.gguf", # llama-2-13b-chat.Q4_0.gguf 파일 경로를 설정한다.

n_gpu_layers = gpuLayerCount,

n_batch = batchSize,

n_ctx = 2048,

f16_kv = True, # 반드시 True로 설정해야 한다. 그렇지 않으면 몇 번의 호출 후에 문제가 발생하게 된다.

verbose = False

)

# 프롬프트 템플리트를 설정한다.

promptTemplate = hub.pull("rlm/rag-prompt")

# 문서 페이지 컨텐츠를 병합하는 함수를 정의한다.

def mergeDocumentPageContent(documentList):

return "\n\n".join(doc.page_content for doc in documentList)

# 실행 체인을 설정한다.

runnableSequence = (

{"context": vectorStoreRetriever | mergeDocumentPageContent, "question" : RunnablePassthrough()}

| promptTemplate

| llamaCpp

| StrOutputParser()

)

# 질의 응답을 실행한다.

resultString = runnableSequence.invoke("What are the approaches to Task Decomposition?")

print(resultString)

"""

The approaches to task decomposition include breaking down large tasks into smaller subgoals, using task-specific instructions, and using human inputs.

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
asgiref==3.8.1
async-timeout==4.0.3
attrs==23.2.0
backoff==2.2.1
bcrypt==4.1.3
beautifulsoup4==4.12.3
bs4==0.0.2
build==1.2.1
cachetools==5.3.3
certifi==2024.6.2
charset-normalizer==3.3.2
chroma-hnswlib==0.7.3
chromadb==0.5.0
click==8.1.7
coloredlogs==15.0.1
dataclasses-json==0.6.7
Deprecated==1.2.14
diskcache==5.6.3
dnspython==2.6.1
email_validator==2.1.1
exceptiongroup==1.2.1
fastapi==0.111.0
fastapi-cli==0.0.4
filelock==3.15.1
flatbuffers==24.3.25
frozenlist==1.4.1
fsspec==2024.6.0
google-auth==2.30.0
googleapis-common-protos==1.63.1
greenlet==3.0.3
grpcio==1.64.1
h11==0.14.0
httpcore==1.0.5
httptools==0.6.1
httpx==0.27.0
huggingface-hub==0.23.3
humanfriendly==10.0
idna==3.7
importlib_metadata==7.1.0
importlib_resources==6.4.0
Jinja2==3.1.4
joblib==1.4.2
jsonpatch==1.33
jsonpointer==3.0.0
kubernetes==30.1.0
langchain==0.2.4
langchain-chroma==0.1.1
langchain-community==0.2.4
langchain-core==0.2.6
langchain-huggingface==0.0.3
langchain-text-splitters==0.2.1
langsmith==0.1.77
llama_cpp_python==0.2.78
markdown-it-py==3.0.0
MarkupSafe==2.1.5
marshmallow==3.21.3
mdurl==0.1.2
mmh3==4.1.0
monotonic==1.6
mpmath==1.3.0
multidict==6.0.5
mypy-extensions==1.0.0
networkx==3.3
numpy==1.26.4
nvidia-cublas-cu12==12.1.3.1
nvidia-cuda-cupti-cu12==12.1.105
nvidia-cuda-nvrtc-cu12==12.1.105
nvidia-cuda-runtime-cu12==12.1.105
nvidia-cudnn-cu12==8.9.2.26
nvidia-cufft-cu12==11.0.2.54
nvidia-curand-cu12==10.3.2.106
nvidia-cusolver-cu12==11.4.5.107
nvidia-cusparse-cu12==12.1.0.106
nvidia-nccl-cu12==2.20.5
nvidia-nvjitlink-cu12==12.5.40
nvidia-nvtx-cu12==12.1.105
oauthlib==3.2.2
onnxruntime==1.18.0
opentelemetry-api==1.25.0
opentelemetry-exporter-otlp-proto-common==1.25.0
opentelemetry-exporter-otlp-proto-grpc==1.25.0
opentelemetry-instrumentation==0.46b0
opentelemetry-instrumentation-asgi==0.46b0
opentelemetry-instrumentation-fastapi==0.46b0
opentelemetry-proto==1.25.0
opentelemetry-sdk==1.25.0
opentelemetry-semantic-conventions==0.46b0
opentelemetry-util-http==0.46b0
orjson==3.10.5
overrides==7.7.0
packaging==24.1
pillow==10.3.0
posthog==3.5.0
protobuf==4.25.3
pyasn1==0.6.0
pyasn1_modules==0.4.0
pydantic==2.7.4
pydantic_core==2.18.4
Pygments==2.18.0
PyPika==0.48.9
pyproject_hooks==1.1.0
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
python-multipart==0.0.9
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
requests-oauthlib==2.0.0
rich==13.7.1
rsa==4.9
safetensors==0.4.3
scikit-learn==1.5.0
scipy==1.13.1
sentence-transformers==3.0.1
shellingham==1.5.4
six==1.16.0
sniffio==1.3.1
soupsieve==2.5
SQLAlchemy==2.0.30
starlette==0.37.2
sympy==1.12.1
tenacity==8.3.0
threadpoolctl==3.5.0
tokenizers==0.19.1
tomli==2.0.1
torch==2.3.1
tqdm==4.66.4
transformers==4.41.2
triton==2.3.1
typer==0.12.3
typing-inspect==0.9.0
typing_extensions==4.12.2
ujson==5.10.0
urllib3==2.2.1
uvicorn==0.30.1
uvloop==0.19.0
watchfiles==0.22.0
websocket-client==1.8.0
websockets==12.0
wrapt==1.16.0
yarl==1.9.4
zipp==3.19.2

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

asgiref==3.8.1

async-timeout==4.0.3

attrs==23.2.0

backoff==2.2.1

bcrypt==4.1.3

beautifulsoup4==4.12.3

bs4==0.0.2

build==1.2.1

cachetools==5.3.3

certifi==2024.6.2

charset-normalizer==3.3.2

chroma-hnswlib==0.7.3

chromadb==0.5.0

click==8.1.7

coloredlogs==15.0.1

dataclasses-json==0.6.7

Deprecated==1.2.14

diskcache==5.6.3

dnspython==2.6.1

email_validator==2.1.1

exceptiongroup==1.2.1

fastapi==0.111.0

fastapi-cli==0.0.4

filelock==3.15.1

flatbuffers==24.3.25

frozenlist==1.4.1

fsspec==2024.6.0

google-auth==2.30.0

googleapis-common-protos==1.63.1

greenlet==3.0.3

grpcio==1.64.1

h11==0.14.0

httpcore==1.0.5

httptools==0.6.1

httpx==0.27.0

huggingface-hub==0.23.3

humanfriendly==10.0

idna==3.7

importlib_metadata==7.1.0

importlib_resources==6.4.0

Jinja2==3.1.4

joblib==1.4.2

jsonpatch==1.33

jsonpointer==3.0.0

kubernetes==30.1.0

langchain==0.2.4

langchain-chroma==0.1.1

langchain-community==0.2.4

langchain-core==0.2.6

langchain-huggingface==0.0.3

langchain-text-splitters==0.2.1

langsmith==0.1.77

llama_cpp_python==0.2.78

markdown-it-py==3.0.0

MarkupSafe==2.1.5

marshmallow==3.21.3

mdurl==0.1.2

mmh3==4.1.0

monotonic==1.6

mpmath==1.3.0

multidict==6.0.5

mypy-extensions==1.0.0

networkx==3.3

numpy==1.26.4

nvidia-cublas-cu12==12.1.3.1

nvidia-cuda-cupti-cu12==12.1.105

nvidia-cuda-nvrtc-cu12==12.1.105

nvidia-cuda-runtime-cu12==12.1.105

nvidia-cudnn-cu12==8.9.2.26

nvidia-cufft-cu12==11.0.2.54

nvidia-curand-cu12==10.3.2.106

nvidia-cusolver-cu12==11.4.5.107

nvidia-cusparse-cu12==12.1.0.106

nvidia-nccl-cu12==2.20.5

nvidia-nvjitlink-cu12==12.5.40

nvidia-nvtx-cu12==12.1.105

oauthlib==3.2.2

onnxruntime==1.18.0

opentelemetry-api==1.25.0

opentelemetry-exporter-otlp-proto-common==1.25.0

opentelemetry-exporter-otlp-proto-grpc==1.25.0

opentelemetry-instrumentation==0.46b0

opentelemetry-instrumentation-asgi==0.46b0

opentelemetry-instrumentation-fastapi==0.46b0

opentelemetry-proto==1.25.0

opentelemetry-sdk==1.25.0

opentelemetry-semantic-conventions==0.46b0

opentelemetry-util-http==0.46b0

orjson==3.10.5

overrides==7.7.0

packaging==24.1

pillow==10.3.0

posthog==3.5.0

protobuf==4.25.3

pyasn1==0.6.0

pyasn1_modules==0.4.0

pydantic==2.7.4

pydantic_core==2.18.4

Pygments==2.18.0

PyPika==0.48.9

pyproject_hooks==1.1.0

python-dateutil==2.9.0.post0

python-dotenv==1.0.1

python-multipart==0.0.9

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

requests-oauthlib==2.0.0

rich==13.7.1

rsa==4.9

safetensors==0.4.3

scikit-learn==1.5.0

scipy==1.13.1

sentence-transformers==3.0.1

shellingham==1.5.4

six==1.16.0

sniffio==1.3.1

soupsieve==2.5

SQLAlchemy==2.0.30

starlette==0.37.2

sympy==1.12.1

tenacity==8.3.0

threadpoolctl==3.5.0

tokenizers==0.19.1

tomli==2.0.1

torch==2.3.1

tqdm==4.66.4

transformers==4.41.2

triton==2.3.1

typer==0.12.3

typing-inspect==0.9.0

typing_extensions==4.12.2

ujson==5.10.0

urllib3==2.2.1

uvicorn==0.30.1

uvloop==0.19.0

watchfiles==0.22.0

websocket-client==1.8.0

websockets==12.0

wrapt==1.16.0

yarl==1.9.4

zipp==3.19.2

※ pip install langchain

[PYTHON/LANGCHAIN] create_retriever_tool 함수 : FAISS 벡터 스토어 검색 도구 만들기 1

■ create_retriever_tool 함수를 사용해 FAISS 벡터 스토어 검색 도구를 만드는 방법을 보여준다. ▶ main.py


import ast
import os
import re

from langchain_community.utilities    import SQLDatabase
from langchain_community.vectorstores import FAISS
from langchain_openai                 import OpenAIEmbeddings
from langchain.agents.agent_toolkits  import create_retriever_tool

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

# SQLITE 데이터베이스를 초기화한다.
sqlDatabase = SQLDatabase.from_uri("sqlite:///chinook.db")

# SQLite 데이터를 가져오는 함수를 정의한다.
def getSQLiteData(sqlDatabase, sql):
    resultString    = sqlDatabase.run(sql)
    resultTupleList = ast.literal_eval(resultString)
    resultList1     = [element for resultTuple in resultTupleList for element in resultTuple if element]
    resultList2     = [re.sub(r"\b\d+\b", "", result).strip() for result in resultList1]
    return list(set(resultList2))

# 아티스트명, 앨범 타이틀 리스트를 구한다.
artistNameList = getSQLiteData(sqlDatabase, "SELECT Name  FROM artists")
albumTitleList = getSQLiteData(sqlDatabase, "SELECT Title FROM albums" )

# FAISS 벡터 스토어를 생성한다:
faiss = FAISS.from_texts(artistNameList + albumTitleList, OpenAIEmbeddings())

# FAISS 벡터 스토어 검색기를 설정한다.
faissVectorStoreRetriever = faiss.as_retriever(search_kwargs = {"k" : 5})

# FAISS 벡터 스토어 검색기 도구를 생성한다.
faissVectorStoreRetrieverTool = create_retriever_tool(
    faissVectorStoreRetriever,
    name        = "search_proper_nouns",
    description = "Use to look up values to filter on. Input is an approximate spelling of the proper noun, output is valid proper nouns. Use the noun most similar to the search."
)

# FAISS 벡터 스토어 검색기 도구를 실행한다.
resultString = faissVectorStoreRetrieverTool.invoke("Alice Chains")

print(resultString)

"""
Alice In Chains

Alanis Morissette

Pearl Jam

Pearl Jam

Audioslave
"""

import ast

import os

import re

from langchain_community.utilities import SQLDatabase

from langchain_community.vectorstores import FAISS

from langchain_openai import OpenAIEmbeddings

from langchain.agents.agent_toolkits import create_retriever_tool

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

# SQLITE 데이터베이스를 초기화한다.

sqlDatabase = SQLDatabase.from_uri("sqlite:///chinook.db")

# SQLite 데이터를 가져오는 함수를 정의한다.

def getSQLiteData(sqlDatabase, sql):

resultString = sqlDatabase.run(sql)

resultTupleList = ast.literal_eval(resultString)

resultList1 = [element for resultTuple in resultTupleList for element in resultTuple if element]

resultList2 = [re.sub(r"\b\d+\b", "", result).strip() for result in resultList1]

return list(set(resultList2))

# 아티스트명, 앨범 타이틀 리스트를 구한다.

artistNameList = getSQLiteData(sqlDatabase, "SELECT Name FROM artists")

albumTitleList = getSQLiteData(sqlDatabase, "SELECT Title FROM albums" )

# FAISS 벡터 스토어를 생성한다:

faiss = FAISS.from_texts(artistNameList + albumTitleList, OpenAIEmbeddings())

# FAISS 벡터 스토어 검색기를 설정한다.

faissVectorStoreRetriever = faiss.as_retriever(search_kwargs = {"k" : 5})

# FAISS 벡터 스토어 검색기 도구를 생성한다.

faissVectorStoreRetrieverTool = create_retriever_tool(

faissVectorStoreRetriever,

name = "search_proper_nouns",

description = "Use to look up values to filter on. Input is an approximate spelling of the proper noun, output is valid proper nouns. Use the noun most similar to the search."

)

# FAISS 벡터 스토어 검색기 도구를 실행한다.

resultString = faissVectorStoreRetrieverTool.invoke("Alice Chains")

print(resultString)

"""

Alice In Chains

Alanis Morissette

Pearl Jam

Audioslave

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.3
langchain-community==0.2.4
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.77
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.33.0
orjson==3.10.4
packaging==23.2
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.3.0
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.1
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.3

langchain-community==0.2.4

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.77

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.33.0

orjson==3.10.4

packaging==23.2

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.3.0

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.1

yarl==1.9.4

※ pip install langchain

[PYTHON/LANGCHAIN] FAISS 클래스 : as_retriever 메소드에서 search_kwargs 인자를 사용해 벡터 스토어 검색기 구하기

■ FAISS 클래스의 as_retriever 메소드에서 search_kwargs 인자를 사용해 벡터 스토어 검색기를 구하는 방법을 보여준다. ※ 검색기가 반환하는 문서 수를 제한할 수도 있다.

[PYTHON/LANGCHAIN] from_texts 함수 : FAISS 벡터 스토어 생성하기

■ from_texts 함수를 사용해 FAISS 벡터 스토어를 생성하는 방법을 보여준다. ▶ main.py


import ast
import os
import re

from langchain_community.utilities    import SQLDatabase
from langchain_community.vectorstores import FAISS
from langchain_openai                 import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

# SQLITE 데이터베이스를 초기화한다.
sqlDatabase = SQLDatabase.from_uri("sqlite:///chinook.db")

# SQLite 데이터를 가져오는 함수를 정의한다.
def getSQLiteData(sqlDatabase, sql):
    resultString    = sqlDatabase.run(sql)
    resultTupleList = ast.literal_eval(resultString)
    resultList1     = [element for resultTuple in resultTupleList for element in resultTuple if element]
    resultList2     = [re.sub(r"\b\d+\b", "", result).strip() for result in resultList1]
    return list(set(resultList2))

# 아티스트명, 앨범 타이틀 리스트를 구한다.
artistNameList = getSQLiteData(sqlDatabase, "SELECT Name  FROM artists")
albumTitleList = getSQLiteData(sqlDatabase, "SELECT Title FROM albums" )

# FAISS 벡터 스토어를 생성한다:
faiss = FAISS.from_texts(artistNameList + albumTitleList, OpenAIEmbeddings())

import ast

import os

import re

from langchain_community.utilities import SQLDatabase

from langchain_community.vectorstores import FAISS

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

# SQLITE 데이터베이스를 초기화한다.

sqlDatabase = SQLDatabase.from_uri("sqlite:///chinook.db")

# SQLite 데이터를 가져오는 함수를 정의한다.

def getSQLiteData(sqlDatabase, sql):

resultString = sqlDatabase.run(sql)

resultTupleList = ast.literal_eval(resultString)

resultList1 = [element for resultTuple in resultTupleList for element in resultTuple if element]

resultList2 = [re.sub(r"\b\d+\b", "", result).strip() for result in resultList1]

return list(set(resultList2))

# 아티스트명, 앨범 타이틀 리스트를 구한다.

artistNameList = getSQLiteData(sqlDatabase, "SELECT Name FROM artists")

albumTitleList = getSQLiteData(sqlDatabase, "SELECT Title FROM albums" )

# FAISS 벡터 스토어를 생성한다:

faiss = FAISS.from_texts(artistNameList + albumTitleList, OpenAIEmbeddings())

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
async-timeout==4.0.3
attrs==23.2.0
certifi==2024.6.2
charset-normalizer==3.3.2
dataclasses-json==0.6.7
distro==1.9.0
exceptiongroup==1.2.1
faiss-gpu==1.7.2
frozenlist==1.4.1
greenlet==3.0.3
h11==0.14.0
httpcore==1.0.5
httpx==0.27.0
idna==3.7
jsonpatch==1.33
jsonpointer==3.0.0
langchain==0.2.3
langchain-community==0.2.4
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.77
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.33.0
orjson==3.10.4
packaging==23.2
pydantic==2.7.4
pydantic_core==2.18.4
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.30
tenacity==8.3.0
tiktoken==0.7.0
tqdm==4.66.4
typing-inspect==0.9.0
typing_extensions==4.12.2
urllib3==2.2.1
yarl==1.9.4

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

async-timeout==4.0.3

attrs==23.2.0

certifi==2024.6.2

charset-normalizer==3.3.2

dataclasses-json==0.6.7

distro==1.9.0

exceptiongroup==1.2.1

faiss-gpu==1.7.2

frozenlist==1.4.1

greenlet==3.0.3

h11==0.14.0

httpcore==1.0.5

httpx==0.27.0

idna==3.7

jsonpatch==1.33

jsonpointer==3.0.0

langchain==0.2.3

langchain-community==0.2.4

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.77

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.33.0

orjson==3.10.4

packaging==23.2

pydantic==2.7.4

pydantic_core==2.18.4

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.30

tenacity==8.3.0

tiktoken==0.7.0

tqdm==4.66.4

typing-inspect==0.9.0

typing_extensions==4.12.2

urllib3==2.2.1

yarl==1.9.4

※ pip install langchain langchain-community langchain-openai

[PYTHON/LANGCHAIN] Chroma 클래스 : as_retriever 메소드를 사용해 VectorStoreRetriever 객체 만들기 1

■ Chroma 클래스의 as_retriever 메소드를 사용해 VectorStoreRetriever 객체를 만드는 방법을 보여준다. ▶ main.py


import os
import bs4

from langchain_community.document_loaders import WebBaseLoader
from langchain_text_splitters             import RecursiveCharacterTextSplitter
from langchain_chroma                     import Chroma
from langchain_openai                     import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

webBaseLoader = WebBaseLoader(
    web_paths = ("https://lilianweng.github.io/posts/2023-06-23-agent/",),
    bs_kwargs = dict(parse_only = bs4.SoupStrainer(class_ = ("post-content", "post-title", "post-header")))
)

documentList = webBaseLoader.load()

recursiveCharacterTextSplitter = RecursiveCharacterTextSplitter(chunk_size = 1000, chunk_overlap = 200)

splitDocumentList = recursiveCharacterTextSplitter.split_documents(documentList)

chroma = Chroma.from_documents(documents = splitDocumentList, embedding = OpenAIEmbeddings())

vectorStoreRetriever = chroma.as_retriever()

import os

import bs4

from langchain_community.document_loaders import WebBaseLoader

from langchain_text_splitters import RecursiveCharacterTextSplitter

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

webBaseLoader = WebBaseLoader(

web_paths = ("https://lilianweng.github.io/posts/2023-06-23-agent/",),

bs_kwargs = dict(parse_only = bs4.SoupStrainer(class_ = ("post-content", "post-title", "post-header")))

)

documentList = webBaseLoader.load()

recursiveCharacterTextSplitter = RecursiveCharacterTextSplitter(chunk_size = 1000, chunk_overlap = 200)

splitDocumentList = recursiveCharacterTextSplitter.split_documents(documentList)

chroma = Chroma.from_documents(documents = splitDocumentList, embedding = OpenAIEmbeddings())

vectorStoreRetriever = chroma.as_retriever()

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
asgiref==3.8.1
async-timeout==4.0.3
attrs==23.2.0
backoff==2.2.1
bcrypt==4.1.3
beautifulsoup4==4.12.3
bs4==0.0.2
build==1.2.1
cachetools==5.3.3
certifi==2024.6.2
charset-normalizer==3.3.2
chroma-hnswlib==0.7.3
chromadb==0.5.0
click==8.1.7
coloredlogs==15.0.1
dataclasses-json==0.6.7
Deprecated==1.2.14
distro==1.9.0
dnspython==2.6.1
email_validator==2.1.1
exceptiongroup==1.2.1
fastapi==0.111.0
fastapi-cli==0.0.4
filelock==3.14.0
flatbuffers==24.3.25
frozenlist==1.4.1
fsspec==2024.6.0
google-auth==2.30.0
googleapis-common-protos==1.63.1
greenlet==3.0.3
grpcio==1.64.1
h11==0.14.0
httpcore==1.0.5
httptools==0.6.1
httpx==0.27.0
huggingface-hub==0.23.3
humanfriendly==10.0
idna==3.7
importlib_metadata==7.1.0
importlib_resources==6.4.0
Jinja2==3.1.4
jsonpatch==1.33
jsonpointer==3.0.0
kubernetes==30.1.0
langchain==0.2.3
langchain-chroma==0.1.1
langchain-community==0.2.4
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.77
markdown-it-py==3.0.0
MarkupSafe==2.1.5
marshmallow==3.21.3
mdurl==0.1.2
mmh3==4.1.0
monotonic==1.6
mpmath==1.3.0
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
oauthlib==3.2.2
onnxruntime==1.18.0
openai==1.33.0
opentelemetry-api==1.25.0
opentelemetry-exporter-otlp-proto-common==1.25.0
opentelemetry-exporter-otlp-proto-grpc==1.25.0
opentelemetry-instrumentation==0.46b0
opentelemetry-instrumentation-asgi==0.46b0
opentelemetry-instrumentation-fastapi==0.46b0
opentelemetry-proto==1.25.0
opentelemetry-sdk==1.25.0
opentelemetry-semantic-conventions==0.46b0
opentelemetry-util-http==0.46b0
orjson==3.10.4
overrides==7.7.0
packaging==23.2
posthog==3.5.0
protobuf==4.25.3
pyasn1==0.6.0
pyasn1_modules==0.4.0
pydantic==2.7.3
pydantic_core==2.18.4
Pygments==2.18.0
PyPika==0.48.9
pyproject_hooks==1.1.0
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
python-multipart==0.0.9
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
requests-oauthlib==2.0.0
rich==13.7.1
rsa==4.9
shellingham==1.5.4
six==1.16.0
sniffio==1.3.1
soupsieve==2.5
SQLAlchemy==2.0.30
starlette==0.37.2
sympy==1.12.1
tenacity==8.3.0
tiktoken==0.7.0
tokenizers==0.19.1
tomli==2.0.1
tqdm==4.66.4
typer==0.12.3
typing-inspect==0.9.0
typing_extensions==4.12.2
ujson==5.10.0
urllib3==2.2.1
uvicorn==0.30.1
uvloop==0.19.0
watchfiles==0.22.0
websocket-client==1.8.0
websockets==12.0
wrapt==1.16.0
yarl==1.9.4
zipp==3.19.2

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

asgiref==3.8.1

async-timeout==4.0.3

attrs==23.2.0

backoff==2.2.1

bcrypt==4.1.3

beautifulsoup4==4.12.3

bs4==0.0.2

build==1.2.1

cachetools==5.3.3

certifi==2024.6.2

charset-normalizer==3.3.2

chroma-hnswlib==0.7.3

chromadb==0.5.0

click==8.1.7

coloredlogs==15.0.1

dataclasses-json==0.6.7

Deprecated==1.2.14

distro==1.9.0

dnspython==2.6.1

email_validator==2.1.1

exceptiongroup==1.2.1

fastapi==0.111.0

fastapi-cli==0.0.4

filelock==3.14.0

flatbuffers==24.3.25

frozenlist==1.4.1

fsspec==2024.6.0

google-auth==2.30.0

googleapis-common-protos==1.63.1

greenlet==3.0.3

grpcio==1.64.1

h11==0.14.0

httpcore==1.0.5

httptools==0.6.1

httpx==0.27.0

huggingface-hub==0.23.3

humanfriendly==10.0

idna==3.7

importlib_metadata==7.1.0

importlib_resources==6.4.0

Jinja2==3.1.4

jsonpatch==1.33

jsonpointer==3.0.0

kubernetes==30.1.0

langchain==0.2.3

langchain-chroma==0.1.1

langchain-community==0.2.4

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.77

markdown-it-py==3.0.0

MarkupSafe==2.1.5

marshmallow==3.21.3

mdurl==0.1.2

mmh3==4.1.0

monotonic==1.6

mpmath==1.3.0

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

oauthlib==3.2.2

onnxruntime==1.18.0

openai==1.33.0

opentelemetry-api==1.25.0

opentelemetry-exporter-otlp-proto-common==1.25.0

opentelemetry-exporter-otlp-proto-grpc==1.25.0

opentelemetry-instrumentation==0.46b0

opentelemetry-instrumentation-asgi==0.46b0

opentelemetry-instrumentation-fastapi==0.46b0

opentelemetry-proto==1.25.0

opentelemetry-sdk==1.25.0

opentelemetry-semantic-conventions==0.46b0

opentelemetry-util-http==0.46b0

orjson==3.10.4

overrides==7.7.0

packaging==23.2

posthog==3.5.0

protobuf==4.25.3

pyasn1==0.6.0

pyasn1_modules==0.4.0

pydantic==2.7.3

pydantic_core==2.18.4

Pygments==2.18.0

PyPika==0.48.9

pyproject_hooks==1.1.0

python-dateutil==2.9.0.post0

python-dotenv==1.0.1

python-multipart==0.0.9

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

requests-oauthlib==2.0.0

rich==13.7.1

rsa==4.9

shellingham==1.5.4

six==1.16.0

sniffio==1.3.1

soupsieve==2.5

SQLAlchemy==2.0.30

starlette==0.37.2

sympy==1.12.1

tenacity==8.3.0

tiktoken==0.7.0

tokenizers==0.19.1

tomli==2.0.1

tqdm==4.66.4

typer==0.12.3

typing-inspect==0.9.0

typing_extensions==4.12.2

ujson==5.10.0

urllib3==2.2.1

uvicorn==0.30.1

uvloop==0.19.0

watchfiles==0.22.0

websocket-client==1.8.0

websockets==12.0

wrapt==1.16.0

yarl==1.9.4

zipp==3.19.2

※ pip install langchain langchain-chroma

[PYTHON/LANGCHAIN] Chroma 클래스 : as_retriever 메소드를 사용해 VectorStoreRetriever 객체 만들기 2

■ Chroma 클래스의 as_retriever 메소드를 사용해 VectorStoreRetriever 객체를 만드는 방법을 보여준다. ▶ main.py


import os

from langchain_core.documents import Document
from langchain_chroma         import Chroma
from langchain_openai         import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [
    Document(
        page_content = "Dogs are great companions, known for their loyalty and friendliness.",
        metadata     = {"source" : "mammal-pets-doc"}
    ),
    Document(
        page_content = "Cats are independent pets that often enjoy their own space.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
    Document(
        page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",
        metadata     = {"source" : "fish-pets-doc"},
    ),
    Document(
        page_content = "Parrots are intelligent birds capable of mimicking human speech.",
        metadata     = {"source" : "bird-pets-doc"},
    ),
    Document(
        page_content = "Rabbits are social animals that need plenty of space to hop around.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
]

chroma = Chroma.from_documents(
    documentList,
    embedding = OpenAIEmbeddings(),
)

vectorStoreRetriever = chroma.as_retriever(
    search_type   = "similarity", # "similarity"(디폴트), "mmr" (maximum marginal relevance, described above), "similarity_score_threshold"
    search_kwargs = {"k" : 1},
)

import os

from langchain_core.documents import Document

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [

Document(

page_content = "Dogs are great companions, known for their loyalty and friendliness.",

metadata = {"source" : "mammal-pets-doc"}

Document(

page_content = "Cats are independent pets that often enjoy their own space.",

metadata = {"source" : "mammal-pets-doc"},

Document(

page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",

metadata = {"source" : "fish-pets-doc"},

Document(

page_content = "Parrots are intelligent birds capable of mimicking human speech.",

metadata = {"source" : "bird-pets-doc"},

Document(

page_content = "Rabbits are social animals that need plenty of space to hop around.",

metadata = {"source" : "mammal-pets-doc"},

]

chroma = Chroma.from_documents(

documentList,

embedding = OpenAIEmbeddings(),

)

vectorStoreRetriever = chroma.as_retriever(

search_type = "similarity", # "similarity"(디폴트), "mmr" (maximum marginal relevance, described above), "similarity_score_threshold"

search_kwargs = {"k" : 1},

)

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
asgiref==3.8.1
async-timeout==4.0.3
attrs==23.2.0
backoff==2.2.1
bcrypt==4.1.3
build==1.2.1
cachetools==5.3.3
certifi==2024.6.2
charset-normalizer==3.3.2
chroma-hnswlib==0.7.3
chromadb==0.5.0
click==8.1.7
coloredlogs==15.0.1
Deprecated==1.2.14
distro==1.9.0
dnspython==2.6.1
email_validator==2.1.1
exceptiongroup==1.2.1
fastapi==0.111.0
fastapi-cli==0.0.4
filelock==3.14.0
flatbuffers==24.3.25
frozenlist==1.4.1
fsspec==2024.6.0
google-auth==2.30.0
googleapis-common-protos==1.63.1
greenlet==3.0.3
grpcio==1.64.1
h11==0.14.0
httpcore==1.0.5
httptools==0.6.1
httpx==0.27.0
huggingface-hub==0.23.3
humanfriendly==10.0
idna==3.7
importlib_metadata==7.1.0
importlib_resources==6.4.0
Jinja2==3.1.4
jsonpatch==1.33
jsonpointer==2.4
kubernetes==30.1.0
langchain==0.2.3
langchain-chroma==0.1.1
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.75
markdown-it-py==3.0.0
MarkupSafe==2.1.5
mdurl==0.1.2
mmh3==4.1.0
monotonic==1.6
mpmath==1.3.0
multidict==6.0.5
numpy==1.26.4
oauthlib==3.2.2
onnxruntime==1.18.0
openai==1.33.0
opentelemetry-api==1.25.0
opentelemetry-exporter-otlp-proto-common==1.25.0
opentelemetry-exporter-otlp-proto-grpc==1.25.0
opentelemetry-instrumentation==0.46b0
opentelemetry-instrumentation-asgi==0.46b0
opentelemetry-instrumentation-fastapi==0.46b0
opentelemetry-proto==1.25.0
opentelemetry-sdk==1.25.0
opentelemetry-semantic-conventions==0.46b0
opentelemetry-util-http==0.46b0
orjson==3.10.3
overrides==7.7.0
packaging==23.2
posthog==3.5.0
protobuf==4.25.3
pyasn1==0.6.0
pyasn1_modules==0.4.0
pydantic==2.7.3
pydantic_core==2.18.4
Pygments==2.18.0
PyPika==0.48.9
pyproject_hooks==1.1.0
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
python-multipart==0.0.9
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
requests-oauthlib==2.0.0
rich==13.7.1
rsa==4.9
shellingham==1.5.4
six==1.16.0
sniffio==1.3.1
SQLAlchemy==2.0.30
starlette==0.37.2
sympy==1.12.1
tenacity==8.3.0
tiktoken==0.7.0
tokenizers==0.19.1
tomli==2.0.1
tqdm==4.66.4
typer==0.12.3
typing_extensions==4.12.2
ujson==5.10.0
urllib3==2.2.1
uvicorn==0.30.1
uvloop==0.19.0
watchfiles==0.22.0
websocket-client==1.8.0
websockets==12.0
wrapt==1.16.0
yarl==1.9.4
zipp==3.19.2

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

asgiref==3.8.1

async-timeout==4.0.3

attrs==23.2.0

backoff==2.2.1

bcrypt==4.1.3

build==1.2.1

cachetools==5.3.3

certifi==2024.6.2

charset-normalizer==3.3.2

chroma-hnswlib==0.7.3

chromadb==0.5.0

click==8.1.7

coloredlogs==15.0.1

Deprecated==1.2.14

distro==1.9.0

dnspython==2.6.1

email_validator==2.1.1

exceptiongroup==1.2.1

fastapi==0.111.0

fastapi-cli==0.0.4

filelock==3.14.0

flatbuffers==24.3.25

frozenlist==1.4.1

fsspec==2024.6.0

google-auth==2.30.0

googleapis-common-protos==1.63.1

greenlet==3.0.3

grpcio==1.64.1

h11==0.14.0

httpcore==1.0.5

httptools==0.6.1

httpx==0.27.0

huggingface-hub==0.23.3

humanfriendly==10.0

idna==3.7

importlib_metadata==7.1.0

importlib_resources==6.4.0

Jinja2==3.1.4

jsonpatch==1.33

jsonpointer==2.4

kubernetes==30.1.0

langchain==0.2.3

langchain-chroma==0.1.1

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.75

markdown-it-py==3.0.0

MarkupSafe==2.1.5

mdurl==0.1.2

mmh3==4.1.0

monotonic==1.6

mpmath==1.3.0

multidict==6.0.5

numpy==1.26.4

oauthlib==3.2.2

onnxruntime==1.18.0

openai==1.33.0

opentelemetry-api==1.25.0

opentelemetry-exporter-otlp-proto-common==1.25.0

opentelemetry-exporter-otlp-proto-grpc==1.25.0

opentelemetry-instrumentation==0.46b0

opentelemetry-instrumentation-asgi==0.46b0

opentelemetry-instrumentation-fastapi==0.46b0

opentelemetry-proto==1.25.0

opentelemetry-sdk==1.25.0

opentelemetry-semantic-conventions==0.46b0

opentelemetry-util-http==0.46b0

orjson==3.10.3

overrides==7.7.0

packaging==23.2

posthog==3.5.0

protobuf==4.25.3

pyasn1==0.6.0

pyasn1_modules==0.4.0

pydantic==2.7.3

pydantic_core==2.18.4

Pygments==2.18.0

PyPika==0.48.9

pyproject_hooks==1.1.0

python-dateutil==2.9.0.post0

python-dotenv==1.0.1

python-multipart==0.0.9

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

requests-oauthlib==2.0.0

rich==13.7.1

rsa==4.9

shellingham==1.5.4

six==1.16.0

sniffio==1.3.1

SQLAlchemy==2.0.30

starlette==0.37.2

sympy==1.12.1

tenacity==8.3.0

tiktoken==0.7.0

tokenizers==0.19.1

tomli==2.0.1

tqdm==4.66.4

typer==0.12.3

typing_extensions==4.12.2

ujson==5.10.0

urllib3==2.2.1

uvicorn==0.30.1

uvloop==0.19.0

watchfiles==0.22.0

websocket-client==1.8.0

websockets==12.0

wrapt==1.16.0

yarl==1.9.4

zipp==3.19.2

※ pip install langchain langchain-chroma

[PYTHON/LANGCHAIN] Chroma 클래스 : similarity_search_by_vector 메소드를 사용해 검색 문자열 벡터 리스트로 검색 결과 문서 리스트 구하기 1

■ Chroma 클래스의 similarity_search_by_vector 메소드를 사용해 검색 문자열의 벡터 리스트로 검색 결과 문서 리스트를 구하는 방법을 보여준다. ▶ main.py


import os

from langchain_core.documents import Document
from langchain_chroma         import Chroma
from langchain_openai         import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [
    Document(
        page_content = "Dogs are great companions, known for their loyalty and friendliness.",
        metadata     = {"source" : "mammal-pets-doc"}
    ),
    Document(
        page_content = "Cats are independent pets that often enjoy their own space.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
    Document(
        page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",
        metadata     = {"source" : "fish-pets-doc"},
    ),
    Document(
        page_content = "Parrots are intelligent birds capable of mimicking human speech.",
        metadata     = {"source" : "bird-pets-doc"},
    ),
    Document(
        page_content = "Rabbits are social animals that need plenty of space to hop around.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
]

openAIEmbeddings = OpenAIEmbeddings()

chroma = Chroma.from_documents(
    documentList,
    embedding = openAIEmbeddings,
)

catEnbeddingList = openAIEmbeddings.embed_query("cat") # catEnbeddingList 항목 수 : 1536개

searchResultList = chroma.similarity_search_by_vector(catEnbeddingList)

print(searchResultList)

"""
[
    Document(
        page_content = 'Cats are independent pets that often enjoy their own space.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Dogs are great companions, known for their loyalty and friendliness.',
        metadata = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Rabbits are social animals that need plenty of space to hop around.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Parrots are intelligent birds capable of mimicking human speech.',
        metadata     = {'source' : 'bird-pets-doc'}
    )
]
"""

import os

from langchain_core.documents import Document

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [

Document(

page_content = "Dogs are great companions, known for their loyalty and friendliness.",

metadata = {"source" : "mammal-pets-doc"}

Document(

page_content = "Cats are independent pets that often enjoy their own space.",

metadata = {"source" : "mammal-pets-doc"},

Document(

page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",

metadata = {"source" : "fish-pets-doc"},

Document(

page_content = "Parrots are intelligent birds capable of mimicking human speech.",

metadata = {"source" : "bird-pets-doc"},

Document(

page_content = "Rabbits are social animals that need plenty of space to hop around.",

metadata = {"source" : "mammal-pets-doc"},

]

openAIEmbeddings = OpenAIEmbeddings()

chroma = Chroma.from_documents(

documentList,

embedding = openAIEmbeddings,

)

catEnbeddingList = openAIEmbeddings.embed_query("cat") # catEnbeddingList 항목 수 : 1536개

searchResultList = chroma.similarity_search_by_vector(catEnbeddingList)

print(searchResultList)

"""

[

Document(

page_content = 'Cats are independent pets that often enjoy their own space.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Dogs are great companions, known for their loyalty and friendliness.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Rabbits are social animals that need plenty of space to hop around.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Parrots are intelligent birds capable of mimicking human speech.',

metadata = {'source' : 'bird-pets-doc'}

)

]

"""

▶ requirements.txt

[PYTHON/LANGCHAIN] Chroma 클래스 : similarity_search_with_score 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트 구하기 (점수 포함)

■ Chroma 클래스의 similarity_search_with_score 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트를 구하는 방법을 보여준다. (점수 포함) ▶ main.py


import os

from langchain_core.documents import Document
from langchain_chroma         import Chroma
from langchain_openai         import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [
    Document(
        page_content = "Dogs are great companions, known for their loyalty and friendliness.",
        metadata     = {"source" : "mammal-pets-doc"}
    ),
    Document(
        page_content = "Cats are independent pets that often enjoy their own space.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
    Document(
        page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",
        metadata     = {"source" : "fish-pets-doc"},
    ),
    Document(
        page_content = "Parrots are intelligent birds capable of mimicking human speech.",
        metadata     = {"source" : "bird-pets-doc"},
    ),
    Document(
        page_content = "Rabbits are social animals that need plenty of space to hop around.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
]

chroma = Chroma.from_documents(
    documentList,
    embedding = OpenAIEmbeddings(),
)

searchResultList = chroma.similarity_search_with_score("cat")

print(searchResultList)

"""
[
    (
        Document(
            page_content = 'Cats are independent pets that often enjoy their own space.',
            metadata     = {'source' : 'mammal-pets-doc'}
        ),
        0.375326931476593
    ),
    (
        Document(
            page_content = 'Dogs are great companions, known for their loyalty and friendliness.',
            metadata     = {'source' : 'mammal-pets-doc'}
        ),
        0.4833090305328369
    ),
    (
        Document(
            page_content = 'Rabbits are social animals that need plenty of space to hop around.',
            metadata     = {'source' : 'mammal-pets-doc'}
        ),
        0.4958883225917816
    ),
    (
        Document(
            page_content = 'Parrots are intelligent birds capable of mimicking human speech.',
            metadata     = {'source' : 'bird-pets-doc'}
        ),
        0.4974174499511719
    )
]
"""

import os

from langchain_core.documents import Document

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [

Document(

page_content = "Dogs are great companions, known for their loyalty and friendliness.",

metadata = {"source" : "mammal-pets-doc"}

Document(

page_content = "Cats are independent pets that often enjoy their own space.",

metadata = {"source" : "mammal-pets-doc"},

Document(

page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",

metadata = {"source" : "fish-pets-doc"},

Document(

page_content = "Parrots are intelligent birds capable of mimicking human speech.",

metadata = {"source" : "bird-pets-doc"},

Document(

page_content = "Rabbits are social animals that need plenty of space to hop around.",

metadata = {"source" : "mammal-pets-doc"},

]

chroma = Chroma.from_documents(

documentList,

embedding = OpenAIEmbeddings(),

)

searchResultList = chroma.similarity_search_with_score("cat")

print(searchResultList)

"""

[

(

Document(

page_content = 'Cats are independent pets that often enjoy their own space.',

metadata = {'source' : 'mammal-pets-doc'}

0.375326931476593

(

Document(

page_content = 'Dogs are great companions, known for their loyalty and friendliness.',

metadata = {'source' : 'mammal-pets-doc'}

0.4833090305328369

(

Document(

page_content = 'Rabbits are social animals that need plenty of space to hop around.',

metadata = {'source' : 'mammal-pets-doc'}

0.4958883225917816

(

Document(

page_content = 'Parrots are intelligent birds capable of mimicking human speech.',

metadata = {'source' : 'bird-pets-doc'}

0.4974174499511719

)

]

"""

▶ requirements.txt

[PYTHON/LANGCHAIN] Chroma 클래스 : asimilarity_search 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트 구하기 (비동기)

■ Chroma 클래스의 asimilarity_search 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트를 구하는 방법을 보여준다. (비동기) ▶ main.py


import asyncio
import os

from langchain_core.documents import Document
from langchain_chroma         import Chroma
from langchain_openai         import OpenAIEmbeddings

async def main():
    os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

    documentList = [
        Document(
            page_content = "Dogs are great companions, known for their loyalty and friendliness.",
            metadata     = {"source" : "mammal-pets-doc"}
        ),
        Document(
            page_content = "Cats are independent pets that often enjoy their own space.",
            metadata     = {"source" : "mammal-pets-doc"},
        ),
        Document(
            page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",
            metadata     = {"source" : "fish-pets-doc"},
        ),
        Document(
            page_content = "Parrots are intelligent birds capable of mimicking human speech.",
            metadata     = {"source" : "bird-pets-doc"},
        ),
        Document(
            page_content = "Rabbits are social animals that need plenty of space to hop around.",
            metadata     = {"source" : "mammal-pets-doc"},
        ),
    ]

    chroma = Chroma.from_documents(
        documentList,
        embedding = OpenAIEmbeddings(),
    )

    searchResultList = await chroma.asimilarity_search("cat")

    print(searchResultList)

asyncio.run(main())

"""
[
    Document(
        page_content = 'Cats are independent pets that often enjoy their own space.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Dogs are great companions, known for their loyalty and friendliness.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Rabbits are social animals that need plenty of space to hop around.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Parrots are intelligent birds capable of mimicking human speech.',
        metadata     = {'source' : 'bird-pets-doc'}
    )
]
"""

import asyncio

import os

from langchain_core.documents import Document

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

async def main():

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [

Document(

page_content = "Dogs are great companions, known for their loyalty and friendliness.",

metadata = {"source" : "mammal-pets-doc"}

Document(

page_content = "Cats are independent pets that often enjoy their own space.",

metadata = {"source" : "mammal-pets-doc"},

Document(

page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",

metadata = {"source" : "fish-pets-doc"},

Document(

page_content = "Parrots are intelligent birds capable of mimicking human speech.",

metadata = {"source" : "bird-pets-doc"},

Document(

page_content = "Rabbits are social animals that need plenty of space to hop around.",

metadata = {"source" : "mammal-pets-doc"},

]

chroma = Chroma.from_documents(

documentList,

embedding = OpenAIEmbeddings(),

)

searchResultList = await chroma.asimilarity_search("cat")

print(searchResultList)

asyncio.run(main())

"""

[

Document(

page_content = 'Cats are independent pets that often enjoy their own space.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Dogs are great companions, known for their loyalty and friendliness.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Rabbits are social animals that need plenty of space to hop around.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Parrots are intelligent birds capable of mimicking human speech.',

metadata = {'source' : 'bird-pets-doc'}

)

]

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
asgiref==3.8.1
async-timeout==4.0.3
attrs==23.2.0
backoff==2.2.1
bcrypt==4.1.3
build==1.2.1
cachetools==5.3.3
certifi==2024.6.2
charset-normalizer==3.3.2
chroma-hnswlib==0.7.3
chromadb==0.5.0
click==8.1.7
coloredlogs==15.0.1
Deprecated==1.2.14
distro==1.9.0
dnspython==2.6.1
email_validator==2.1.1
exceptiongroup==1.2.1
fastapi==0.111.0
fastapi-cli==0.0.4
filelock==3.14.0
flatbuffers==24.3.25
frozenlist==1.4.1
fsspec==2024.6.0
google-auth==2.30.0
googleapis-common-protos==1.63.1
greenlet==3.0.3
grpcio==1.64.1
h11==0.14.0
httpcore==1.0.5
httptools==0.6.1
httpx==0.27.0
huggingface-hub==0.23.3
humanfriendly==10.0
idna==3.7
importlib_metadata==7.1.0
importlib_resources==6.4.0
Jinja2==3.1.4
jsonpatch==1.33
jsonpointer==2.4
kubernetes==30.1.0
langchain==0.2.3
langchain-chroma==0.1.1
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.75
markdown-it-py==3.0.0
MarkupSafe==2.1.5
mdurl==0.1.2
mmh3==4.1.0
monotonic==1.6
mpmath==1.3.0
multidict==6.0.5
numpy==1.26.4
oauthlib==3.2.2
onnxruntime==1.18.0
openai==1.33.0
opentelemetry-api==1.25.0
opentelemetry-exporter-otlp-proto-common==1.25.0
opentelemetry-exporter-otlp-proto-grpc==1.25.0
opentelemetry-instrumentation==0.46b0
opentelemetry-instrumentation-asgi==0.46b0
opentelemetry-instrumentation-fastapi==0.46b0
opentelemetry-proto==1.25.0
opentelemetry-sdk==1.25.0
opentelemetry-semantic-conventions==0.46b0
opentelemetry-util-http==0.46b0
orjson==3.10.3
overrides==7.7.0
packaging==23.2
posthog==3.5.0
protobuf==4.25.3
pyasn1==0.6.0
pyasn1_modules==0.4.0
pydantic==2.7.3
pydantic_core==2.18.4
Pygments==2.18.0
PyPika==0.48.9
pyproject_hooks==1.1.0
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
python-multipart==0.0.9
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
requests-oauthlib==2.0.0
rich==13.7.1
rsa==4.9
shellingham==1.5.4
six==1.16.0
sniffio==1.3.1
SQLAlchemy==2.0.30
starlette==0.37.2
sympy==1.12.1
tenacity==8.3.0
tiktoken==0.7.0
tokenizers==0.19.1
tomli==2.0.1
tqdm==4.66.4
typer==0.12.3
typing_extensions==4.12.2
ujson==5.10.0
urllib3==2.2.1
uvicorn==0.30.1
uvloop==0.19.0
watchfiles==0.22.0
websocket-client==1.8.0
websockets==12.0
wrapt==1.16.0
yarl==1.9.4
zipp==3.19.2

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

asgiref==3.8.1

async-timeout==4.0.3

attrs==23.2.0

backoff==2.2.1

bcrypt==4.1.3

build==1.2.1

cachetools==5.3.3

certifi==2024.6.2

charset-normalizer==3.3.2

chroma-hnswlib==0.7.3

chromadb==0.5.0

click==8.1.7

coloredlogs==15.0.1

Deprecated==1.2.14

distro==1.9.0

dnspython==2.6.1

email_validator==2.1.1

exceptiongroup==1.2.1

fastapi==0.111.0

fastapi-cli==0.0.4

filelock==3.14.0

flatbuffers==24.3.25

frozenlist==1.4.1

fsspec==2024.6.0

google-auth==2.30.0

googleapis-common-protos==1.63.1

greenlet==3.0.3

grpcio==1.64.1

h11==0.14.0

httpcore==1.0.5

httptools==0.6.1

httpx==0.27.0

huggingface-hub==0.23.3

humanfriendly==10.0

idna==3.7

importlib_metadata==7.1.0

importlib_resources==6.4.0

Jinja2==3.1.4

jsonpatch==1.33

jsonpointer==2.4

kubernetes==30.1.0

langchain==0.2.3

langchain-chroma==0.1.1

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.75

markdown-it-py==3.0.0

MarkupSafe==2.1.5

mdurl==0.1.2

mmh3==4.1.0

monotonic==1.6

mpmath==1.3.0

multidict==6.0.5

numpy==1.26.4

oauthlib==3.2.2

onnxruntime==1.18.0

openai==1.33.0

opentelemetry-api==1.25.0

opentelemetry-exporter-otlp-proto-common==1.25.0

opentelemetry-exporter-otlp-proto-grpc==1.25.0

opentelemetry-instrumentation==0.46b0

opentelemetry-instrumentation-asgi==0.46b0

opentelemetry-instrumentation-fastapi==0.46b0

opentelemetry-proto==1.25.0

opentelemetry-sdk==1.25.0

opentelemetry-semantic-conventions==0.46b0

opentelemetry-util-http==0.46b0

orjson==3.10.3

overrides==7.7.0

packaging==23.2

posthog==3.5.0

protobuf==4.25.3

pyasn1==0.6.0

pyasn1_modules==0.4.0

pydantic==2.7.3

pydantic_core==2.18.4

Pygments==2.18.0

PyPika==0.48.9

pyproject_hooks==1.1.0

python-dateutil==2.9.0.post0

python-dotenv==1.0.1

python-multipart==0.0.9

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

requests-oauthlib==2.0.0

rich==13.7.1

rsa==4.9

shellingham==1.5.4

six==1.16.0

sniffio==1.3.1

SQLAlchemy==2.0.30

starlette==0.37.2

sympy==1.12.1

tenacity==8.3.0

tiktoken==0.7.0

tokenizers==0.19.1

tomli==2.0.1

tqdm==4.66.4

typer==0.12.3

typing_extensions==4.12.2

ujson==5.10.0

urllib3==2.2.1

uvicorn==0.30.1

uvloop==0.19.0

watchfiles==0.22.0

websocket-client==1.8.0

websockets==12.0

wrapt==1.16.0

yarl==1.9.4

zipp==3.19.2

[PYTHON/LANGCHAIN] Chroma 클래스 : similarity_search 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트 구하기

■ Chroma 클래스의 similarity_search 메소드를 사용해 검색 문자열로 검색 결과 문서 리스트를 구하는 방법을 보여준다. ▶ main.py


import os

from langchain_core.documents import Document
from langchain_chroma         import Chroma
from langchain_openai         import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [
    Document(
        page_content = "Dogs are great companions, known for their loyalty and friendliness.",
        metadata     = {"source" : "mammal-pets-doc"}
    ),
    Document(
        page_content = "Cats are independent pets that often enjoy their own space.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
    Document(
        page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",
        metadata     = {"source" : "fish-pets-doc"},
    ),
    Document(
        page_content = "Parrots are intelligent birds capable of mimicking human speech.",
        metadata     = {"source" : "bird-pets-doc"},
    ),
    Document(
        page_content = "Rabbits are social animals that need plenty of space to hop around.",
        metadata     = {"source" : "mammal-pets-doc"},
    ),
]

chroma = Chroma.from_documents(
)
    documentList,
    embedding = OpenAIEmbeddings(),

searchResultList = chroma.similarity_search("cat")

print(searchResultList)

"""
[
    Document(
        page_content = 'Cats are independent pets that often enjoy their own space.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Dogs are great companions, known for their loyalty and friendliness.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Rabbits are social animals that need plenty of space to hop around.',
        metadata     = {'source' : 'mammal-pets-doc'}
    ),
    Document(
        page_content = 'Parrots are intelligent birds capable of mimicking human speech.',
        metadata     = {'source' : 'bird-pets-doc'}
    )
]
"""

import os

from langchain_core.documents import Document

from langchain_chroma import Chroma

from langchain_openai import OpenAIEmbeddings

os.environ["OPENAI_API_KEY"] = "<OPENAI_API_KEY>"

documentList = [

Document(

page_content = "Dogs are great companions, known for their loyalty and friendliness.",

metadata = {"source" : "mammal-pets-doc"}

Document(

page_content = "Cats are independent pets that often enjoy their own space.",

metadata = {"source" : "mammal-pets-doc"},

Document(

page_content = "Goldfish are popular pets for beginners, requiring relatively simple care.",

metadata = {"source" : "fish-pets-doc"},

Document(

page_content = "Parrots are intelligent birds capable of mimicking human speech.",

metadata = {"source" : "bird-pets-doc"},

Document(

page_content = "Rabbits are social animals that need plenty of space to hop around.",

metadata = {"source" : "mammal-pets-doc"},

]

chroma = Chroma.from_documents(

)

documentList,

embedding = OpenAIEmbeddings(),

searchResultList = chroma.similarity_search("cat")

print(searchResultList)

"""

[

Document(

page_content = 'Cats are independent pets that often enjoy their own space.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Dogs are great companions, known for their loyalty and friendliness.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Rabbits are social animals that need plenty of space to hop around.',

metadata = {'source' : 'mammal-pets-doc'}

Document(

page_content = 'Parrots are intelligent birds capable of mimicking human speech.',

metadata = {'source' : 'bird-pets-doc'}

)

]

"""

▶ requirements.txt


aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.7.0
anyio==4.4.0
asgiref==3.8.1
async-timeout==4.0.3
attrs==23.2.0
backoff==2.2.1
bcrypt==4.1.3
build==1.2.1
cachetools==5.3.3
certifi==2024.6.2
charset-normalizer==3.3.2
chroma-hnswlib==0.7.3
chromadb==0.5.0
click==8.1.7
coloredlogs==15.0.1
Deprecated==1.2.14
distro==1.9.0
dnspython==2.6.1
email_validator==2.1.1
exceptiongroup==1.2.1
fastapi==0.111.0
fastapi-cli==0.0.4
filelock==3.14.0
flatbuffers==24.3.25
frozenlist==1.4.1
fsspec==2024.6.0
google-auth==2.30.0
googleapis-common-protos==1.63.1
greenlet==3.0.3
grpcio==1.64.1
h11==0.14.0
httpcore==1.0.5
httptools==0.6.1
httpx==0.27.0
huggingface-hub==0.23.3
humanfriendly==10.0
idna==3.7
importlib_metadata==7.1.0
importlib_resources==6.4.0
Jinja2==3.1.4
jsonpatch==1.33
jsonpointer==2.4
kubernetes==30.1.0
langchain==0.2.3
langchain-chroma==0.1.1
langchain-core==0.2.5
langchain-openai==0.1.8
langchain-text-splitters==0.2.1
langsmith==0.1.75
markdown-it-py==3.0.0
MarkupSafe==2.1.5
mdurl==0.1.2
mmh3==4.1.0
monotonic==1.6
mpmath==1.3.0
multidict==6.0.5
numpy==1.26.4
oauthlib==3.2.2
onnxruntime==1.18.0
openai==1.33.0
opentelemetry-api==1.25.0
opentelemetry-exporter-otlp-proto-common==1.25.0
opentelemetry-exporter-otlp-proto-grpc==1.25.0
opentelemetry-instrumentation==0.46b0
opentelemetry-instrumentation-asgi==0.46b0
opentelemetry-instrumentation-fastapi==0.46b0
opentelemetry-proto==1.25.0
opentelemetry-sdk==1.25.0
opentelemetry-semantic-conventions==0.46b0
opentelemetry-util-http==0.46b0
orjson==3.10.3
overrides==7.7.0
packaging==23.2
posthog==3.5.0
protobuf==4.25.3
pyasn1==0.6.0
pyasn1_modules==0.4.0
pydantic==2.7.3
pydantic_core==2.18.4
Pygments==2.18.0
PyPika==0.48.9
pyproject_hooks==1.1.0
python-dateutil==2.9.0.post0
python-dotenv==1.0.1
python-multipart==0.0.9
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
requests-oauthlib==2.0.0
rich==13.7.1
rsa==4.9
shellingham==1.5.4
six==1.16.0
sniffio==1.3.1
SQLAlchemy==2.0.30
starlette==0.37.2
sympy==1.12.1
tenacity==8.3.0
tiktoken==0.7.0
tokenizers==0.19.1
tomli==2.0.1
tqdm==4.66.4
typer==0.12.3
typing_extensions==4.12.2
ujson==5.10.0
urllib3==2.2.1
uvicorn==0.30.1
uvloop==0.19.0
watchfiles==0.22.0
websocket-client==1.8.0
websockets==12.0
wrapt==1.16.0
yarl==1.9.4
zipp==3.19.2

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

aiohttp==3.9.5

aiosignal==1.3.1

annotated-types==0.7.0

anyio==4.4.0

asgiref==3.8.1

async-timeout==4.0.3

attrs==23.2.0

backoff==2.2.1

bcrypt==4.1.3

build==1.2.1

cachetools==5.3.3

certifi==2024.6.2

charset-normalizer==3.3.2

chroma-hnswlib==0.7.3

chromadb==0.5.0

click==8.1.7

coloredlogs==15.0.1

Deprecated==1.2.14

distro==1.9.0

dnspython==2.6.1

email_validator==2.1.1

exceptiongroup==1.2.1

fastapi==0.111.0

fastapi-cli==0.0.4

filelock==3.14.0

flatbuffers==24.3.25

frozenlist==1.4.1

fsspec==2024.6.0

google-auth==2.30.0

googleapis-common-protos==1.63.1

greenlet==3.0.3

grpcio==1.64.1

h11==0.14.0

httpcore==1.0.5

httptools==0.6.1

httpx==0.27.0

huggingface-hub==0.23.3

humanfriendly==10.0

idna==3.7

importlib_metadata==7.1.0

importlib_resources==6.4.0

Jinja2==3.1.4

jsonpatch==1.33

jsonpointer==2.4

kubernetes==30.1.0

langchain==0.2.3

langchain-chroma==0.1.1

langchain-core==0.2.5

langchain-openai==0.1.8

langchain-text-splitters==0.2.1

langsmith==0.1.75

markdown-it-py==3.0.0

MarkupSafe==2.1.5

mdurl==0.1.2

mmh3==4.1.0

monotonic==1.6

mpmath==1.3.0

multidict==6.0.5

numpy==1.26.4

oauthlib==3.2.2

onnxruntime==1.18.0

openai==1.33.0

opentelemetry-api==1.25.0

opentelemetry-exporter-otlp-proto-common==1.25.0

opentelemetry-exporter-otlp-proto-grpc==1.25.0

opentelemetry-instrumentation==0.46b0

opentelemetry-instrumentation-asgi==0.46b0

opentelemetry-instrumentation-fastapi==0.46b0

opentelemetry-proto==1.25.0

opentelemetry-sdk==1.25.0

opentelemetry-semantic-conventions==0.46b0

opentelemetry-util-http==0.46b0

orjson==3.10.3

overrides==7.7.0

packaging==23.2

posthog==3.5.0

protobuf==4.25.3

pyasn1==0.6.0

pyasn1_modules==0.4.0

pydantic==2.7.3

pydantic_core==2.18.4

Pygments==2.18.0

PyPika==0.48.9

pyproject_hooks==1.1.0

python-dateutil==2.9.0.post0

python-dotenv==1.0.1

python-multipart==0.0.9

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

requests-oauthlib==2.0.0

rich==13.7.1

rsa==4.9

shellingham==1.5.4

six==1.16.0

sniffio==1.3.1

SQLAlchemy==2.0.30

starlette==0.37.2

sympy==1.12.1

tenacity==8.3.0

tiktoken==0.7.0

tokenizers==0.19.1

tomli==2.0.1

tqdm==4.66.4

typer==0.12.3

typing_extensions==4.12.2

ujson==5.10.0

urllib3==2.2.1

uvicorn==0.30.1

uvloop==0.19.0

watchfiles==0.22.0

websocket-client==1.8.0

websockets==12.0

wrapt==1.16.0

yarl==1.9.4

zipp==3.19.2

※