[PYTHON/LANGCHAIN] MaxMarginalRelevanceExampleSelector 클래스 : 최대 주변 관련성(MMR, Maximal Marginal Relevance) 예제 선택기 설정하기

■ MaxMarginalRelevanceExampleSelector 클래스를 사용해 최대 주변 관련성(MMR, Maximal Marginal Relevance) 예제 선택기를 설정하는 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_core.example_selectors import MaxMarginalRelevanceExampleSelector
from langchain_core.prompts           import FewShotPromptTemplate
from langchain_core.prompts           import PromptTemplate
from langchain_openai                 import OpenAIEmbeddings

load_dotenv()

examplePromptTemplate = PromptTemplate(
    input_variables = ["input", "output"],
    template        = "Input : {input}\nOutput : {output}"
)

exampleDictionaryList = [
    {"input" : "happy"    , "output" : "sad"      },
    {"input" : "tall"     , "output" : "short"    },
    {"input" : "energetic", "output" : "lethargic"},
    {"input" : "sunny"    , "output" : "gloomy"   },
    {"input" : "windy"    , "output" : "calm"     }
]

maxMarginalRelevanceExampleSelector = MaxMarginalRelevanceExampleSelector.from_examples(
    # 선택할 수 있는 예제 목록이다.
    exampleDictionaryList,
    # 의미론적 유사성을 측정하는 데 사용되는 임베딩을 생성하는 데 사용되는 임베딩 클래스이다.
    OpenAIEmbeddings(),
    # 임베딩을 저장하고 유사성 검색을 수행하는 데 사용되는 VectorStore 클래스이다.
    FAISS,
    # 생성할 예제 수이다.
    k = 2
)

fewShotPromptTemplate = FewShotPromptTemplate(
    # 예제 대신 exampleSelector를 제공한다.
    example_selector = maxMarginalRelevanceExampleSelector,
    example_prompt   = examplePromptTemplate,
    prefix           = "Give the antonym of every input",
    suffix           = "Input : {adjective}\nOutput :",
    input_variables  = ["adjective"]
)

promptString = fewShotPromptTemplate.format(adjective = "worried")

print(promptString)

"""
Give the antonym of every input

Input : happy
Output : sad

Input : windy
Output : calm

Input : worried
Output :
"""

from dotenv import load_dotenv

from langchain_community.vectorstores import FAISS

from langchain_core.example_selectors import MaxMarginalRelevanceExampleSelector

from langchain_core.prompts import FewShotPromptTemplate

from langchain_core.prompts import PromptTemplate

from langchain_openai import OpenAIEmbeddings

load_dotenv()

examplePromptTemplate = PromptTemplate(

input_variables = ["input", "output"],

template = "Input : {input}\nOutput : {output}"

)

exampleDictionaryList = [

{"input" : "happy" , "output" : "sad" },

{"input" : "tall" , "output" : "short" },

{"input" : "energetic", "output" : "lethargic"},

{"input" : "sunny" , "output" : "gloomy" },

{"input" : "windy" , "output" : "calm" }

]

maxMarginalRelevanceExampleSelector = MaxMarginalRelevanceExampleSelector.from_examples(

# 선택할 수 있는 예제 목록이다.

exampleDictionaryList,

# 의미론적 유사성을 측정하는 데 사용되는 임베딩을 생성하는 데 사용되는 임베딩 클래스이다.

OpenAIEmbeddings(),

# 임베딩을 저장하고 유사성 검색을 수행하는 데 사용되는 VectorStore 클래스이다.

FAISS,

# 생성할 예제 수이다.

k = 2

)

fewShotPromptTemplate = FewShotPromptTemplate(

# 예제 대신 exampleSelector를 제공한다.

example_selector = maxMarginalRelevanceExampleSelector,

example_prompt = examplePromptTemplate,

prefix = "Give the antonym of every input",

suffix = "Input : {adjective}\nOutput :",

input_variables = ["adjective"]

)

promptString = fewShotPromptTemplate.format(adjective = "worried")

print(promptString)

"""

Give the antonym of every input

Input : happy

Output : sad

Input : windy

Output : calm

Input : worried

Output :

"""

▶ 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==0.2.9
langchain-openai==0.1.9
langchain-text-splitters==0.2.1
langsmith==0.1.81
marshmallow==3.21.3
multidict==6.0.5
mypy-extensions==1.0.0
numpy==1.26.4
openai==1.35.3
orjson==3.10.5
packaging==24.1
pydantic==2.7.4
pydantic_core==2.18.4
python-dotenv==1.0.1
PyYAML==6.0.1
regex==2024.5.15
requests==2.32.3
sniffio==1.3.1
SQLAlchemy==2.0.31
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==0.2.9

langchain-openai==0.1.9

langchain-text-splitters==0.2.1

langsmith==0.1.81

marshmallow==3.21.3

multidict==6.0.5

mypy-extensions==1.0.0

numpy==1.26.4

openai==1.35.3

orjson==3.10.5

packaging==24.1

pydantic==2.7.4

pydantic_core==2.18.4

python-dotenv==1.0.1

PyYAML==6.0.1

regex==2024.5.15

requests==2.32.3

sniffio==1.3.1

SQLAlchemy==2.0.31

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 python-dotenv langchain langchain-community langchain-openai faiss-gpu 명령을 실행했다.