■ PipelinePromptTemplate 클래스를 사용해 파이프라인 프롬프트 템플리트를 만드는 방법을 보여준다.
▶ main.py
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 |
from langchain_core.prompts import PromptTemplate from langchain_core.prompts import PipelinePromptTemplate introductionPromptTemplateString = """You are impersonating {person}.""" introductionPromptTemplate = PromptTemplate.from_template(introductionPromptTemplateString) examplePromptTemplateString = """Here's an example of an interaction: Q : {example_question} A : {example_answer}""" examplePromptTemplate = PromptTemplate.from_template(examplePromptTemplateString) startPromptTemplateString = """Now, do this for real! Q : {input} A :""" startPromptTemplate = PromptTemplate.from_template(startPromptTemplateString) inputPromptTemplateTupleList = [ ("introduction", introductionPromptTemplate), ("example" , examplePromptTemplate ), ("start" , startPromptTemplate ) ] fullPromptTemplateString = """{introduction} {example} {start}""" fullPromptTemplate = PromptTemplate.from_template(fullPromptTemplateString) pipelinePromptTemplate = PipelinePromptTemplate(final_prompt = fullPromptTemplate, pipeline_prompts = inputPromptTemplateTupleList) |
▶ requirements.txt
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 |
aiohttp==3.9.5 aiosignal==1.3.1 annotated-types==0.7.0 async-timeout==4.0.3 attrs==23.2.0 certifi==2024.6.2 charset-normalizer==3.3.2 frozenlist==1.4.1 greenlet==3.0.3 idna==3.7 jsonpatch==1.33 jsonpointer==3.0.0 langchain==0.2.5 langchain-core==0.2.9 langchain-text-splitters==0.2.1 langsmith==0.1.81 multidict==6.0.5 numpy==1.26.4 orjson==3.10.5 packaging==24.1 pydantic==2.7.4 pydantic_core==2.18.4 PyYAML==6.0.1 requests==2.32.3 SQLAlchemy==2.0.31 tenacity==8.4.1 typing_extensions==4.12.2 urllib3==2.2.2 yarl==1.9.4 |
※ pip install langchain 명령을 실행했다.