학습 내용
이 가이드에서는 다음 방법을 설명합니다.- 지식 베이스 구축하기
- 관련 문서를 찾는 검색 step이 포함된 RAG 애플리케이션 만들기
- Weave로 검색 step 추적하기
- LLM 평가자로 RAG 애플리케이션을 평가하여 컨텍스트 정밀도 측정하기
- 맞춤형 점수화 함수 정의하기
사전 요구 사항
- CoreWeave Forge 계정
- Python 3.10 이상 또는 Node.js 18 이상
- 필수 패키지 설치:
- Python:
pip install weave openai - TypeScript:
npm install weave openai
- Python:
- 환경 변수로 설정된 OpenAI API 키
지식 베이스 구축하기
지식 베이스는 RAG 애플리케이션이 쿼리 시점에 검색하는 말뭉치입니다. 이 섹션에서는 소수의 아티클에 대한 벡터 임베딩을 계산해 두고, 이후 주어진 질문과 가장 관련성이 높은 아티클을 검색할 수 있도록 합니다. 먼저 아티클의 임베딩을 계산합니다. 보통은 이 작업을 한 번만 수행한 뒤 임베딩과 메타데이터를 데이터베이스에 저장하지만, 여기서는 간단하게 하기 위해 스크립트가 실행될 때마다 계산합니다.- Python
- TypeScript
from openai import OpenAI
import weave
from weave import Model
import numpy as np
import json
import asyncio
articles = [
"Novo Nordisk and Eli Lilly rival soars 32 percent after promising weight loss drug results Shares of Denmarks Zealand Pharma shot 32 percent higher in morning trade, after results showed success in its liver disease treatment survodutide, which is also on trial as a drug to treat obesity. The trial “tells us that the 6mg dose is safe, which is the top dose used in the ongoing [Phase 3] obesity trial too,” one analyst said in a note. The results come amid feverish investor interest in drugs that can be used for weight loss.",
"Berkshire shares jump after big profit gain as Buffetts conglomerate nears $1 trillion valuation Berkshire Hathaway shares rose on Monday after Warren Buffetts conglomerate posted strong earnings for the fourth quarter over the weekend. Berkshires Class A and B shares jumped more than 1.5%, each. Class A shares are higher by more than 17% this year, while Class B has gained more than 18%. Berkshire was last valued at $930.1 billion, up from $905.5 billion where it closed on Friday, according to FactSet. Berkshire on Saturday posted fourth-quarter operating earnings of $8.481 billion, about 28 percent higher than the $6.625 billion from the year-ago period, driven by big gains in its insurance business. Operating earnings refers to profits from businesses across insurance, railroads and utilities. Meanwhile, Berkshires cash levels also swelled to record levels. The conglomerate held $167.6 billion in cash in the fourth quarter, surpassing the $157.2 billion record the conglomerate held in the prior quarter.",
"Highmark Health says its combining tech from Google and Epic to give doctors easier access to information Highmark Health announced it is integrating technology from Google Cloud and the health-care software company Epic Systems. The integration aims to make it easier for both payers and providers to access key information they need, even if its stored across multiple points and formats, the company said. Highmark is the parent company of a health plan with 7 million members, a provider network of 14 hospitals and other entities",
"Rivian and Lucid shares plunge after weak EV earnings reports Shares of electric vehicle makers Rivian and Lucid fell Thursday after the companies reported stagnant production in their fourth-quarter earnings after the bell Wednesday. Rivian shares sank about 25 percent, and Lucids stock dropped around 17 percent. Rivian forecast it will make 57,000 vehicles in 2024, slightly less than the 57,232 vehicles it produced in 2023. Lucid said it expects to make 9,000 vehicles in 2024, more than the 8,428 vehicles it made in 2023.",
"Mauritius blocks Norwegian cruise ship over fears of a potential cholera outbreak Local authorities on Sunday denied permission for the Norwegian Dawn ship, which has 2,184 passengers and 1,026 crew on board, to access the Mauritius capital of Port Louis, citing “potential health risks.” The Mauritius Ports Authority said Sunday that samples were taken from at least 15 passengers on board the cruise ship. A spokesperson for the U.S.-headquartered Norwegian Cruise Line Holdings said Sunday that 'a small number of guests experienced mild symptoms of a stomach-related illness' during Norwegian Dawns South Africa voyage.",
"Intuitive Machines lands on the moon in historic first for a U.S. company Intuitive Machines Nova-C cargo lander, named Odysseus after the mythological Greek hero, is the first U.S. spacecraft to soft land on the lunar surface since 1972. Intuitive Machines is the first company to pull off a moon landing — government agencies have carried out all previously successful missions. The company's stock surged in extended trading Thursday, after falling 11 percent in regular trading.",
"Lunar landing photos: Intuitive Machines Odysseus sends back first images from the moon Intuitive Machines cargo moon lander Odysseus returned its first images from the surface. Company executives believe the lander caught its landing gear sideways on the moon's surface while touching down and tipped over. Despite resting on its side, the company's historic IM-1 mission is still operating on the moon.",
]
def docs_to_embeddings(docs: list) -> list:
openai = OpenAI()
document_embeddings = []
for doc in docs:
response = (
openai.embeddings.create(input=doc, model="text-embedding-3-small")
.data[0]
.embedding
)
document_embeddings.append(response)
return document_embeddings
article_embeddings = docs_to_embeddings(articles) # 참고: 일반적으로 이 작업은 아티클에 대해 한 번만 수행하고, 임베딩과 메타데이터는 데이터베이스에 저장해 둡니다
require('dotenv').config();
import { OpenAI } from 'openai';
import * as weave from 'weave';
interface Article {
text: string;
embedding?: number[];
}
const articles: Article[] = [
{
text: `Novo Nordisk and Eli Lilly rival soars 32 percent after promising weight loss drug results Shares of Denmarks Zealand Pharma shot 32 percent higher in morning trade, after results showed success in its liver disease treatment survodutide, which is also on trial as a drug to treat obesity. The trial tells us that the 6mg dose is safe, which is the top dose used in the ongoing [Phase 3] obesity trial too, one analyst said in a note. The results come amid feverish investor interest in drugs that can be used for weight loss.`
},
{
text: `Berkshire shares jump after big profit gain as Buffetts conglomerate nears $1 trillion valuation Berkshire Hathaway shares rose on Monday after Warren Buffetts conglomerate posted strong earnings for the fourth quarter over the weekend. Berkshires Class A and B shares jumped more than 1.5%, each. Class A shares are higher by more than 17% this year, while Class B has gained more than 18%. Berkshire was last valued at $930.1 billion, up from $905.5 billion where it closed on Friday, according to FactSet. Berkshire on Saturday posted fourth-quarter operating earnings of $8.481 billion, about 28 percent higher than the $6.625 billion from the year-ago period, driven by big gains in its insurance business. Operating earnings refers to profits from businesses across insurance, railroads and utilities. Meanwhile, Berkshires cash levels also swelled to record levels. The conglomerate held $167.6 billion in cash in the fourth quarter, surpassing the $157.2 billion record the conglomerate held in the prior quarter.`
},
{
text: `Highmark Health says its combining tech from Google and Epic to give doctors easier access to information Highmark Health announced it is integrating technology from Google Cloud and the health-care software company Epic Systems. The integration aims to make it easier for both payers and providers to access key information they need, even if its stored across multiple points and formats, the company said. Highmark is the parent company of a health plan with 7 million members, a provider network of 14 hospitals and other entities`
}
];
function cosineSimilarity(a: number[], b: number[]): number {
const dotProduct = a.reduce((sum, val, i) => sum + val * b[i], 0);
const magnitudeA = Math.sqrt(a.reduce((sum, val) => sum + val * val, 0));
const magnitudeB = Math.sqrt(b.reduce((sum, val) => sum + val * val, 0));
return dotProduct / (magnitudeA * magnitudeB);
}
const docsToEmbeddings = weave.op(async function(docs: Article[]): Promise<Article[]> {
const openai = new OpenAI();
const enrichedDocs = await Promise.all(docs.map(async (doc) => {
const response = await openai.embeddings.create({
input: doc.text,
model: "text-embedding-3-small"
});
return {
...doc,
embedding: response.data[0].embedding
};
}));
return enrichedDocs;
});
RAG 앱 만들기
지식 베이스가 준비되었으니 이제 RAG 애플리케이션을 구축해 보겠습니다. 이 섹션에서는 검색 step과 LLM Call을 결합하고, 두 단계를 모두 Weave로 래핑하여 모든 입력과 출력이 자동으로 추적되도록 합니다. 다음으로, 검색 함수get_most_relevant_document를 weave.op() 데코레이터로 래핑하고 Model 클래스를 만드세요. 검색 함수를 weave.op()로 래핑하면 Weave가 호출될 때마다 입력과 출력을 캡처하므로, 나중에 검색 step을 확인할 수 있습니다. weave.init('<team-name>/rag-quickstart')를 호출하면 함수의 모든 입력과 출력이 추적되기 시작하며, 이를 나중에 확인할 수 있습니다. 팀 이름을 지정하지 않으면 Weave는 출력을 W&B 기본 팀 또는 entity에 기록합니다.
- Python
- TypeScript
from openai import OpenAI
import weave
from weave import Model
import numpy as np
import asyncio
@weave.op()
def get_most_relevant_document(query):
openai = OpenAI()
query_embedding = (
openai.embeddings.create(input=query, model="text-embedding-3-small")
.data[0]
.embedding
)
similarities = [
np.dot(query_embedding, doc_emb)
/ (np.linalg.norm(query_embedding) * np.linalg.norm(doc_emb))
for doc_emb in article_embeddings
]
# 가장 유사한 문서의 인덱스 조회
most_relevant_doc_index = np.argmax(similarities)
return articles[most_relevant_doc_index]
class RAGModel(Model):
system_message: str
model_name: str = "gpt-3.5-turbo-1106"
@weave.op()
def predict(self, question: str) -> dict: # 참고: `question`은 나중에 평가 행에서 데이터를 선택하는 데 사용됩니다
from openai import OpenAI
context = get_most_relevant_document(question)
client = OpenAI()
query = f"""Use the following information to answer the subsequent question. If the answer cannot be found, write "I don't know."
Context:
\"\"\"
{context}
\"\"\"
Question: {question}"""
response = client.chat.completions.create(
model=self.model_name,
messages=[
{"role": "system", "content": self.system_message},
{"role": "user", "content": query},
],
temperature=0.0,
response_format={"type": "text"},
)
answer = response.choices[0].message.content
return {'answer': answer, 'context': context}
# 팀 및 프로젝트 이름 설정
weave.init('<team-name>/rag-quickstart')
model = RAGModel(
system_message="You are an expert in finance and answer questions related to finance, financial services, and financial markets. When responding based on provided information, be sure to cite the source."
)
model.predict("What significant result was reported about Zealand Pharma's obesity trial?")
class RAGModel {
private openai: OpenAI;
private systemMessage: string;
private modelName: string;
private articleEmbeddings: Article[];
constructor(config: {
systemMessage: string;
modelName?: string;
articleEmbeddings: Article[];
}) {
this.openai = new OpenAI();
this.systemMessage = config.systemMessage;
this.modelName = config.modelName || "gpt-3.5-turbo-1106";
this.articleEmbeddings = config.articleEmbeddings;
this.predict = weave.op(this, this.predict);
}
async predict(question: string): Promise<{
answer: string;
context: string;
}> {
const context = await this.getMostRelevantDocument(question);
const response = await this.openai.chat.completions.create({
model: this.modelName,
messages: [
{ role: "system", content: this.systemMessage },
{ role: "user", content: `Use the following information to answer the subsequent question. If the answer cannot be found, write "I don't know."
Context:
"""
${context}
"""
Question: ${question}` }
],
temperature: 0
});
return {
answer: response.choices[0].message.content || "",
context
};
}
}
LLM 평가자로 평가하기
이제 RAG 애플리케이션이 실행되고 Weave로 추적되고 있으니, 다음 단계는 애플리케이션이 질문에 얼마나 잘 답변하는지 평가하는 것입니다. 이 섹션에서는 LLM을 자동 평가자로 활용해 레이블을 직접 작성하지 않고도 애플리케이션의 응답에 점수를 매기는 방법을 설명합니다. 애플리케이션을 평가할 단순한 방법이 없다면 LLM을 사용해 애플리케이션의 여러 측면을 평가할 수 있습니다. 다음은 LLM 평가자에게 주어진 답변을 도출하는 데 컨텍스트가 유용했는지 확인하도록 프롬프트를 제공해 컨텍스트 정밀도(context precision)를 측정하는 예시입니다. 이 프롬프트는 널리 사용되는 RAGAS 프레임워크의 프롬프트를 보강한 것입니다.점수화 함수 정의하기
Evaluation 파이프라인 구축 튜토리얼에서와 마찬가지로, 앱을 테스트할 예제 행 세트와 점수화 함수를 정의합니다. 점수화 함수는 행 하나를 받아 평가합니다. 입력 인수는 행의 해당 키와 일치해야 하므로, 여기서question은 행 딕셔너리에서 가져옵니다. output은 모델의 출력입니다. 모델의 입력도 입력 인수에 따라 예제에서 가져오므로, 여기서도 question이 사용됩니다. 이 예제에서는 함수가 병렬로 실행되도록 async 함수를 사용합니다. async를 간단히 알아보려면 Python asyncio 문서를 참조하세요.
- Python
- TypeScript
from openai import OpenAI
import weave
import asyncio
@weave.op()
async def context_precision_score(question, output):
context_precision_prompt = """Given question, answer and context verify if the context was useful in arriving at the given answer. Give verdict as "1" if useful and "0" if not with json output.
Output in only valid JSON format.
question: {question}
context: {context}
answer: {answer}
verdict: """
client = OpenAI()
prompt = context_precision_prompt.format(
question=question,
context=output['context'],
answer=output['answer'],
)
response = client.chat.completions.create(
model="gpt-4-turbo-preview",
messages=[{"role": "user", "content": prompt}],
response_format={ "type": "json_object" }
)
response_message = response.choices[0].message
response = json.loads(response_message.content)
return {
"verdict": int(response["verdict"]) == 1,
}
questions = [
{"question": "What significant result was reported about Zealand Pharma's obesity trial?"},
{"question": "How much did Berkshire Hathaway's cash levels increase in the fourth quarter?"},
{"question": "What is the goal of Highmark Health's integration of Google Cloud and Epic Systems technology?"},
{"question": "What were Rivian and Lucid's vehicle production forecasts for 2024?"},
{"question": "Why was the Norwegian Dawn cruise ship denied access to Mauritius?"},
{"question": "Which company achieved the first U.S. moon landing since 1972?"},
{"question": "What issue did Intuitive Machines' lunar lander encounter upon landing on the moon?"}
]
evaluation = weave.Evaluation(dataset=questions, scorers=[context_precision_score])
asyncio.run(evaluation.evaluate(model)) # 참고: 평가할 모델을 정의해야 합니다
const contextPrecisionScore = weave.op(async function(args: {
datasetRow: QuestionRow;
modelOutput: { answer: string; context: string; }
}): Promise<ScorerResult> {
const openai = new OpenAI();
const prompt = `Given question, answer and context verify if the context was useful...`;
const response = await openai.chat.completions.create({
model: "gpt-4-turbo-preview",
messages: [{ role: "user", content: prompt }],
response_format: { type: "json_object" }
});
const result = JSON.parse(response.choices[0].message.content || "{}");
return {
verdict: parseInt(result.verdict) === 1
};
});
const evaluation = new weave.Evaluation({
dataset: createQuestionDataset(),
scorers: [contextPrecisionScore]
});
await evaluation.evaluate({
model: weave.op((args: { datasetRow: QuestionRow }) =>
model.predict(args.datasetRow.question)
)
});
선택: Scorer 클래스 정의하기
이전 섹션의 점수화 함수는 단순한 경우에 적합하지만, 여러 평가에서 동일한 평가자를 재사용하거나 점수 집계 방식을 사용자 지정하려면 Scorer 클래스를 사용하는 것이 좋습니다. 다음 단계에서는 이 클래스가 필요한 경우와 정의하는 방법을 설명합니다.
애플리케이션에 따라 맞춤형 평가 클래스(예: 채팅 모델, 프롬프트 등의 매개변수를 갖는 표준화된 LLMJudge 클래스)나 행별 맞춤형 점수화, 집계 점수의 맞춤형 계산이 필요할 수 있습니다. Weave는 바로 사용할 수 있는 Scorer 클래스 목록을 제공하며, 맞춤형 Scorer도 쉽게 만들 수 있습니다. 다음 예시는 맞춤형 class CorrectnessLLMJudge(Scorer)를 만드는 방법을 보여 줍니다.
맞춤형 Scorer를 만드는 단계는 대략 다음과 같습니다.
weave.flow.scorer.Scorer를 상속하는 맞춤형 클래스를 정의합니다.score함수를 재정의하고, 함수의 모든 호출을 추적하려면@weave.op()를 추가합니다.- 이 함수에는 모델의 예측이 전달되는
output인수를 정의해야 합니다. 모델이 “None”을 반환할 수 있으므로Optional[dict]유형으로 정의하세요. - 나머지 인수는 일반적인
Any또는dict로 정의하거나,weave.Evaluate클래스로 모델을 평가할 때 사용하는 데이터셋의 특정 열을 선택하도록 정의할 수 있습니다. 이러한 인수의 이름은 열 이름과 정확히 일치해야 하며,preprocess_model_input을 사용하는 경우에는 이 함수를 거친 단일 행의 키와 정확히 일치해야 합니다.
- 이 함수에는 모델의 예측이 전달되는
- 선택: 집계 점수 계산 방식을 사용자 지정하려면
summarize함수를 재정의합니다. 맞춤형 함수를 정의하지 않으면 Weave는 기본적으로weave.flow.scorer.auto_summarize함수를 사용합니다.- 이 함수에는
@weave.op()데코레이터를 지정해야 합니다.
- 이 함수에는
- Python
- TypeScript
from weave import Scorer
class CorrectnessLLMJudge(Scorer):
prompt: str
model_name: str
device: str
@weave.op()
async def score(self, output: Optional[dict], query: str, answer: str) -> Any:
"""pred, query, target을 비교하여 예측의 정확성을 점수화합니다.
인수:
- output: 평가 대상 모델이 제공하는 dict
- query: 질문 - 데이터셋에 정의된 값
- answer: 목표 답변 - 데이터셋에 정의된 값
반환값:
- 단일 dict {metric name: 단일 평가 값}"""
# get_model은 제공된 params(OpenAI, HF 등)를 기반으로 모델을 가져오는 범용 getter로 정의됨
eval_model = get_model(
model_name = self.model_name,
prompt = self.prompt
device = self.device,
)
# 평가 속도를 높이기 위한 비동기 평가 - 반드시 비동기일 필요는 없음
grade = await eval_model.async_predict(
{
"query": query,
"answer": answer,
"result": output.get("result"),
}
)
# 출력 파싱 - pydantic을 사용하면 더 견고하게 처리할 수 있음
evaluation = "incorrect" not in grade["text"].strip().lower()
# Weave에 표시되는 열 이름
return {"correct": evaluation}
@weave.op()
def summarize(self, score_rows: list) -> Optional[dict]:
"""점수화 함수가 각 행에 대해 계산한 모든 점수를 집계합니다.
인수:
- score_rows: dict 목록. 각 dict에는 메트릭과 점수가 포함됨
반환값:
- 입력과 동일한 구조의 중첩 dict"""
# 아무것도 제공하지 않으면 weave.flow.scorer.auto_summarize 함수가 사용됨
# return auto_summarize(score_rows)
valid_data = [x.get("correct") for x in score_rows if x.get("correct") is not None]
count_true = list(valid_data).count(True)
int_data = [int(x) for x in valid_data]
sample_mean = np.mean(int_data) if int_data else 0
sample_variance = np.var(int_data) if int_data else 0
sample_error = np.sqrt(sample_variance / len(int_data)) if int_data else 0
# 추가 "correct" 계층은 필수는 아니지만 UI에서 구조를 잡아 줌
return {
"correct": {
"true_count": count_true,
"true_fraction": sample_mean,
"stderr": sample_error,
}
}
이 기능은 아직 TypeScript에서 사용할 수 없습니다.
Evaluation의 scorers 인수에 전달하세요.
- Python
- TypeScript
evaluation = weave.Evaluation(dataset=questions, scorers=[CorrectnessLLMJudge()])
이 기능은 아직 TypeScript에서 사용할 수 없습니다.
전체 통합하기
이 섹션에서는 앞선 단계의 모든 내용을 하나의 엔드투엔드 예시로 통합합니다. 이를 통해 각 구성 요소가 어떻게 맞물리는지 확인하고, 여러분의 RAG 애플리케이션에 맞게 응용할 수 있습니다. 여러분의 RAG 앱에서도 동일한 결과를 얻으려면 다음을 수행하세요.- LLM Call과 검색 step 함수를
weave.op()로 래핑하세요. - 선택:
predict함수와 앱 세부 정보를 포함하는Model서브클래스를 만드세요. - 평가할 예시를 수집하세요.
- 예시 하나에 점수를 매기는 점수화 함수를 만드세요.
Evaluation클래스를 사용하여 예시에 대한 평가를 실행하세요.
WEAVE_PARALLELISM=3).
전체 코드는 다음과 같습니다.
- Python
- TypeScript
from openai import OpenAI
import weave
from weave import Model
import numpy as np
import json
import asyncio
# 평가에 사용할 예시
articles = [
"Novo Nordisk and Eli Lilly rival soars 32 percent after promising weight loss drug results Shares of Denmarks Zealand Pharma shot 32 percent higher in morning trade, after results showed success in its liver disease treatment survodutide, which is also on trial as a drug to treat obesity. The trial “tells us that the 6mg dose is safe, which is the top dose used in the ongoing [Phase 3] obesity trial too,” one analyst said in a note. The results come amid feverish investor interest in drugs that can be used for weight loss.",
"Berkshire shares jump after big profit gain as Buffetts conglomerate nears $1 trillion valuation Berkshire Hathaway shares rose on Monday after Warren Buffetts conglomerate posted strong earnings for the fourth quarter over the weekend. Berkshires Class A and B shares jumped more than 1.5%, each. Class A shares are higher by more than 17% this year, while Class B has gained more than 18%. Berkshire was last valued at $930.1 billion, up from $905.5 billion where it closed on Friday, according to FactSet. Berkshire on Saturday posted fourth-quarter operating earnings of $8.481 billion, about 28 percent higher than the $6.625 billion from the year-ago period, driven by big gains in its insurance business. Operating earnings refers to profits from businesses across insurance, railroads and utilities. Meanwhile, Berkshires cash levels also swelled to record levels. The conglomerate held $167.6 billion in cash in the fourth quarter, surpassing the $157.2 billion record the conglomerate held in the prior quarter.",
"Highmark Health says its combining tech from Google and Epic to give doctors easier access to information Highmark Health announced it is integrating technology from Google Cloud and the health-care software company Epic Systems. The integration aims to make it easier for both payers and providers to access key information they need, even if it's stored across multiple points and formats, the company said. Highmark is the parent company of a health plan with 7 million members, a provider network of 14 hospitals and other entities",
"Rivian and Lucid shares plunge after weak EV earnings reports Shares of electric vehicle makers Rivian and Lucid fell Thursday after the companies reported stagnant production in their fourth-quarter earnings after the bell Wednesday. Rivian shares sank about 25 percent, and Lucids stock dropped around 17 percent. Rivian forecast it will make 57,000 vehicles in 2024, slightly less than the 57,232 vehicles it produced in 2023. Lucid said it expects to make 9,000 vehicles in 2024, more than the 8,428 vehicles it made in 2023.",
"Mauritius blocks Norwegian cruise ship over fears of a potential cholera outbreak Local authorities on Sunday denied permission for the Norwegian Dawn ship, which has 2,184 passengers and 1,026 crew on board, to access the Mauritius capital of Port Louis, citing “potential health risks.” The Mauritius Ports Authority said Sunday that samples were taken from at least 15 passengers on board the cruise ship. A spokesperson for the U.S.-headquartered Norwegian Cruise Line Holdings said Sunday that 'a small number of guests experienced mild symptoms of a stomach-related illness' during Norwegian Dawns South Africa voyage.",
"Intuitive Machines lands on the moon in historic first for a U.S. company Intuitive Machines Nova-C cargo lander, named Odysseus after the mythological Greek hero, is the first U.S. spacecraft to soft land on the lunar surface since 1972. Intuitive Machines is the first company to pull off a moon landing — government agencies have carried out all previously successful missions. The company's stock surged in extended trading Thursday, after falling 11 percent in regular trading.",
"Lunar landing photos: Intuitive Machines Odysseus sends back first images from the moon Intuitive Machines cargo moon lander Odysseus returned its first images from the surface. Company executives believe the lander caught its landing gear sideways on the surface of the moon while touching down and tipped over. Despite resting on its side, the company's historic IM-1 mission is still operating on the moon.",
]
def docs_to_embeddings(docs: list) -> list:
openai = OpenAI()
document_embeddings = []
for doc in docs:
response = (
openai.embeddings.create(input=doc, model="text-embedding-3-small")
.data[0]
.embedding
)
document_embeddings.append(response)
return document_embeddings
article_embeddings = docs_to_embeddings(articles) # 참고: 일반적으로 이 작업은 아티클에 대해 한 번만 수행한 뒤 임베딩과 메타데이터를 데이터베이스에 저장합니다
# 검색 step에 데코레이터 추가
@weave.op()
def get_most_relevant_document(query):
openai = OpenAI()
query_embedding = (
openai.embeddings.create(input=query, model="text-embedding-3-small")
.data[0]
.embedding
)
similarities = [
np.dot(query_embedding, doc_emb)
/ (np.linalg.norm(query_embedding) * np.linalg.norm(doc_emb))
for doc_emb in article_embeddings
]
# 가장 유사한 문서의 인덱스 조회
most_relevant_doc_index = np.argmax(similarities)
return articles[most_relevant_doc_index]
# 앱 세부 정보와 응답을 생성하는 predict 함수를 포함하는 Model 서브클래스 생성
class RAGModel(Model):
system_message: str
model_name: str = "gpt-3.5-turbo-1106"
@weave.op()
def predict(self, question: str) -> dict: # 참고: `question`은 나중에 평가 행에서 데이터를 선택할 때 사용됩니다
from openai import OpenAI
context = get_most_relevant_document(question)
client = OpenAI()
query = f"""Use the following information to answer the subsequent question. If the answer cannot be found, write "I don't know."
Context:
\"\"\"
{context}
\"\"\"
Question: {question}"""
response = client.chat.completions.create(
model=self.model_name,
messages=[
{"role": "system", "content": self.system_message},
{"role": "user", "content": query},
],
temperature=0.0,
response_format={"type": "text"},
)
answer = response.choices[0].message.content
return {'answer': answer, 'context': context}
# 팀 이름과 프로젝트 이름 설정
weave.init('<team-name>/rag-quickstart')
model = RAGModel(
system_message="You are an expert in finance and answer questions related to finance, financial services, and financial markets. When responding based on provided information, be sure to cite the source."
)
# 질문과 출력을 사용해 점수를 산출하는 점수화 함수
@weave.op()
async def context_precision_score(question, output):
context_precision_prompt = """Given question, answer and context verify if the context was useful in arriving at the given answer. Give verdict as "1" if useful and "0" if not with json output.
Output in only valid JSON format.
question: {question}
context: {context}
answer: {answer}
verdict: """
client = OpenAI()
prompt = context_precision_prompt.format(
question=question,
context=output['context'],
answer=output['answer'],
)
response = client.chat.completions.create(
model="gpt-4-turbo-preview",
messages=[{"role": "user", "content": prompt}],
response_format={ "type": "json_object" }
)
response_message = response.choices[0].message
response = json.loads(response_message.content)
return {
"verdict": int(response["verdict"]) == 1,
}
questions = [
{"question": "What significant result was reported about Zealand Pharma's obesity trial?"},
{"question": "How much did Berkshire Hathaway's cash levels increase in the fourth quarter?"},
{"question": "What is the goal of Highmark Health's integration of Google Cloud and Epic Systems technology?"},
{"question": "What were Rivian and Lucid's vehicle production forecasts for 2024?"},
{"question": "Why was the Norwegian Dawn cruise ship denied access to Mauritius?"},
{"question": "Which company achieved the first U.S. moon landing since 1972?"},
{"question": "What issue did Intuitive Machines' lunar lander encounter upon landing on the moon?"}
]
# Evaluation 객체를 정의하고 예시 질문과 점수화 함수를 전달
evaluation = weave.Evaluation(dataset=questions, scorers=[context_precision_score])
asyncio.run(evaluation.evaluate(model))
require('dotenv').config();
import { OpenAI } from 'openai';
import * as weave from 'weave';
interface Article {
text: string;
embedding?: number[];
}
const articles: Article[] = [
{
text: `Novo Nordisk and Eli Lilly rival soars 32 percent after promising weight loss drug results Shares of Denmarks Zealand Pharma shot 32 percent higher in morning trade, after results showed success in its liver disease treatment survodutide, which is also on trial as a drug to treat obesity. The trial tells us that the 6mg dose is safe, which is the top dose used in the ongoing [Phase 3] obesity trial too, one analyst said in a note. The results come amid feverish investor interest in drugs that can be used for weight loss.`
},
{
text: `Berkshire shares jump after big profit gain as Buffetts conglomerate nears $1 trillion valuation Berkshire Hathaway shares rose on Monday after Warren Buffetts conglomerate posted strong earnings for the fourth quarter over the weekend. Berkshires Class A and B shares jumped more than 1.5%, each. Class A shares are higher by more than 17% this year, while Class B has gained more than 18%. Berkshire was last valued at $930.1 billion, up from $905.5 billion where it closed on Friday, according to FactSet. Berkshire on Saturday posted fourth-quarter operating earnings of $8.481 billion, about 28 percent higher than the $6.625 billion from the year-ago period, driven by big gains in its insurance business. Operating earnings refers to profits from businesses across insurance, railroads and utilities. Meanwhile, Berkshires cash levels also swelled to record levels. The conglomerate held $167.6 billion in cash in the fourth quarter, surpassing the $157.2 billion record the conglomerate held in the prior quarter.`
},
{
text: `Highmark Health says its combining tech from Google and Epic to give doctors easier access to information Highmark Health announced it is integrating technology from Google Cloud and the health-care software company Epic Systems. The integration aims to make it easier for both payers and providers to access key information they need, even if its stored across multiple points and formats, the company said. Highmark is the parent company of a health plan with 7 million members, a provider network of 14 hospitals and other entities`
}
];
function cosineSimilarity(a: number[], b: number[]): number {
const dotProduct = a.reduce((sum, val, i) => sum + val * b[i], 0);
const magnitudeA = Math.sqrt(a.reduce((sum, val) => sum + val * val, 0));
const magnitudeB = Math.sqrt(b.reduce((sum, val) => sum + val * val, 0));
return dotProduct / (magnitudeA * magnitudeB);
}
const docsToEmbeddings = weave.op(async function(docs: Article[]): Promise<Article[]> {
const openai = new OpenAI();
const enrichedDocs = await Promise.all(docs.map(async (doc) => {
const response = await openai.embeddings.create({
input: doc.text,
model: "text-embedding-3-small"
});
return {
...doc,
embedding: response.data[0].embedding
};
}));
return enrichedDocs;
});
class RAGModel {
private openai: OpenAI;
private systemMessage: string;
private modelName: string;
private articleEmbeddings: Article[];
constructor(config: {
systemMessage: string;
modelName?: string;
articleEmbeddings: Article[];
}) {
this.openai = new OpenAI();
this.systemMessage = config.systemMessage;
this.modelName = config.modelName || "gpt-3.5-turbo-1106";
this.articleEmbeddings = config.articleEmbeddings;
this.predict = weave.op(this, this.predict);
}
private async getMostRelevantDocument(query: string): Promise<string> {
const queryEmbedding = await this.openai.embeddings.create({
input: query,
model: "text-embedding-3-small"
});
const similarities = this.articleEmbeddings.map(doc => {
if (!doc.embedding) return 0;
return cosineSimilarity(queryEmbedding.data[0].embedding, doc.embedding);
});
const mostRelevantIndex = similarities.indexOf(Math.max(...similarities));
return this.articleEmbeddings[mostRelevantIndex].text;
}
async predict(question: string): Promise<{
answer: string;
context: string;
}> {
const context = await this.getMostRelevantDocument(question);
const response = await this.openai.chat.completions.create({
model: this.modelName,
messages: [
{ role: "system", content: this.systemMessage },
{
role: "user",
content: `Use the following information to answer the subsequent question. If the answer cannot be found, write "I don't know."
Context:
"""
${context}
"""
Question: ${question}`
}
],
temperature: 0
});
return {
answer: response.choices[0].message.content || "",
context
};
}
}
interface ScorerResult {
verdict: boolean;
}
interface QuestionRow {
question: string;
}
function createQuestionDataset(): weave.Dataset<QuestionRow> {
return new weave.Dataset<QuestionRow>({
id: 'rag-questions',
rows: [
{ question: "What significant result was reported about Zealand Pharma's obesity trial?" },
{ question: "How much did Berkshire Hathaway's cash levels increase in the fourth quarter?" },
{ question: "What is the goal of Highmark Health's integration of Google Cloud and Epic Systems technology?" }
]
});
}
const contextPrecisionScore = weave.op(async function(args: {
datasetRow: QuestionRow;
modelOutput: { answer: string; context: string; }
}): Promise<ScorerResult> {
const openai = new OpenAI();
const prompt = `Given question, answer and context verify if the context was useful in arriving at the given answer. Give verdict as "1" if useful and "0" if not with json output.
Output in only valid JSON format.
question: ${args.datasetRow.question}
context: ${args.modelOutput.context}
answer: ${args.modelOutput.answer}
verdict: `;
const response = await openai.chat.completions.create({
model: "gpt-4-turbo-preview",
messages: [{ role: "user", content: prompt }],
response_format: { type: "json_object" }
});
const result = JSON.parse(response.choices[0].message.content || "{}");
return {
verdict: parseInt(result.verdict) === 1
};
});
async function main() {
# 팀 이름과 프로젝트 이름을 설정하세요
await weave.init('<team-name>/rag-quickstart');
const articleEmbeddings = await docsToEmbeddings(articles);
const model = new RAGModel({
systemMessage: "You are an expert in finance and answer questions related to finance, financial services, and financial markets. When responding based on provided information, be sure to cite the source.",
articleEmbeddings
});
const evaluation = new weave.Evaluation({
dataset: createQuestionDataset(),
scorers: [contextPrecisionScore]
});
const results = await evaluation.evaluate({
model: weave.op((args: { datasetRow: QuestionRow }) =>
model.predict(args.datasetRow.question)
)
});
console.log('Evaluation results:', results);
}
if (require.main === module) {
main().catch(console.error);
}