انتقل إلى المحتوى الرئيسي

Ragas

بداية سريعة

المفاهيم

概念含义
Question
ContextsRetrieved contexts:实际找到的Context
Answer最终生成的答案
Ground truths参考答案

مجموعة البيانات

from datasets import Dataset 

data_samples = {
'question': ['When was the first super bowl?', 'Who won the most super bowls?'],
'answer': ['The first superbowl was held on January 15, 1967', 'The most super bowls have been won by The New England Patriots'],
'contexts' : [['The Super Bowl....season since 1966,','replacing the NFL...in February.'],
['The Green Bay Packers...Green Bay, Wisconsin.','The Packers compete...Football Conference']],
'ground_truth': ['The first superbowl was held on January 15, 1967', 'The New England Patriots have won the Super Bowl a record six times']
}

dataset = Dataset.from_dict(data_samples)

متري

evol-generate

官方文档

Metric
Context PrecisionRetrievalQuestion是否跑题:检索结果 与 Quesion 是否相关
Answer RelevanceAnwserQuestion是否跑题:生成的答案 是否与 Question 相关
FaithfulnessAnwserRetrieval是否参考引用:生成的答案 是否忠诚于 检索结果
Context RecallRetrieval参考答案
Ground Truth
检索的准确性: 检索结果 与 参考答案 是否相关

المطلوب

دقة السياق

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.

The output should be a well-formatted JSON instance that conforms to the JSON schema below.

……

Your actual task:

question: 法国的首都是什么?
context: 巴黎是法国的首都。
answer: 巴黎
verification:

أهمية الإجابة

هذا Prompt لا يفهم

Generate a question for the given answer and Identify if answer is noncommittal. Give noncommittal as 1 if the answer is noncommittal and 0 if the answer is committal. A noncommittal answer is one that is evasive, vague, or ambiguous. For example, "I don't know" or "I'm not sure" are noncommittal answers

……

Your actual task:

answer: 巴黎
context: 巴黎是法国的首都。
output:

الأمانة

Create one or more statements from each sentence in the given answer.

……

Your actual task:

question: 法国的首都是什么?
answer: 巴黎
statements:

Your task is to judge the faithfulness of a series of statements based on a given context. For each statement you must return verdict as 1 if the statement can be verified based on the context or 0 if the statement can not be verified based on the context.

……

Your actual task:

context: 巴黎是法国的首都。
statements: ["\u6cd5\u56fd\u7684\u9996\u90fd\u662f\u5df4\u9ece\u3002"]
answer:

استدعاء السياق

Given a context, and an answer, analyze each sentence in the answer and classify if the sentence can be attributed to the given context or not. Use only "Yes" (1) or "No" (0) as a binary classification. Output json with reason.

……

Your actual task:

question: 法国的首都是什么?
context: 巴黎是法国的首都。
answer: 巴黎
classification:

توليد بيانات مركبة

الدافع: إنشاء مئات من عينات QA (السؤال - السياق - الإجابة) يدويًا من المستند قد يستغرق وقتاً طويلاً وشاقة.استخدم LLM لتوليد تلقائيا.

方案:Evol-Instruct

الفئات: simple, reasoning, conditioning, multi-context

هذه الفئة تسمى evolutions

evol-generate

عند إنتاج، تحديد النسب المئوية لهذه الفئات الثلاث.

from ragas.testset.generator import TestsetGenerator
from ragas.testset.evolutions import simple, reasoning, multi_context
from langchain_openai import ChatOpenAI, OpenAIEmbeddings

# documents = load your documents

# generator with openai models
generator_llm = ChatOpenAI(model="gpt-3.5-turbo-16k")
critic_llm = ChatOpenAI(model="gpt-4")
embeddings = OpenAIEmbeddings()

generator = TestsetGenerator.from_langchain(
generator_llm,
critic_llm,
embeddings
)

# Change resulting question type distribution
distributions = {
simple: 0.5,
multi_context: 0.4,
reasoning: 0.1
}

# use generator.generate_with_llamaindex_docs if you use llama-index as document loader
testset = generator.generate_with_langchain_docs(documents, 10, distributions)
testset.to_pandas()

قراءة البيانات

استخدام LangChain الرسمي ، والاستمرار في استخدام LangChain ، ليس من السهل أن يكون هناك مشاكل.

التكيف التلقائي للغة

官网

التقييم

ترجمة Prompt المستخدمة في عملية التقييم إلى اللغة الصينية باستخدام نموذج gpt-4-turbo-preview؛ تخزين مؤقت محليا.

سيتم الآن تكييف المطالبات التي تنتمي إلى المقاييس المعنية تلقائيًا مع اللغة المستهدفة.

تقوم خطوة الحفظ بحفظها في .cacha/ragas بشكل افتراضي لإعادة استخدامها لاحقًا.

# 将Metric中的Prompt翻译成中文
from datasets import Dataset
# from langchain.chat_models import ChatOpenAI
from langchain_openai import ChatOpenAI, OpenAI

from ragas.metrics import (
answer_relevancy,
faithfulness,
context_recall,
context_precision,
answer_correctness,
answer_similarity,
)
from ragas import evaluate
from ragas import adapt

eval_model = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)

# llm used for adaptation
openai_model = ChatOpenAI(model_name="gpt-4-turbo-preview")
# openai_model = OpenAI(model_name="gpt-4-0125-preview", temperature=0)

adapt(
metrics=[
answer_relevancy,
# faithfulness,
context_recall,
context_precision,

answer_correctness,
# answer_similarity,
],
language="Chinese",
llm=openai_model,
)

# Eval
dataset = Dataset.from_dict(
{
"question": ["法国的首都是什么?"],
"contexts": [["巴黎是法国的首都。"]],
"answer": ["巴黎"],
"ground_truths": [["巴黎"]],
}
)
print(dataset)

results = evaluate(dataset, llm=eval_model)
print(results)

البيانات المخزنة المؤقتة المنتجة

image-20240416184615163

توليد بيانات مركبة

from ragas.testset.generator import TestsetGenerator
from ragas.testset.evolutions import simple, reasoning, multi_context,conditional
from langchain_openai import ChatOpenAI, OpenAIEmbeddings

# generator with openai models
generator_llm = ChatOpenAI(model="gpt-3.5-turbo-16k")
critic_llm = ChatOpenAI(model="gpt-4")
embeddings = OpenAIEmbeddings()

generator = TestsetGenerator.from_langchain(
generator_llm,
critic_llm,
embeddings
)

# adapt to language
language = "Chinese"

generator.adapt(language, evolutions=[simple, reasoning,conditional,multi_context])
generator.save(evolutions=[simple, reasoning, multi_context,conditional])

البيانات المخزنة المؤقتة المنتجة

image-20240416184635299

الحقول في dataset

  • السؤال
  • contexts: السياق الذي تم استرجاعه
  • ground_truth: إجابة مرجعية
  • anwser: إجابات تم إنشاؤها
  1. Question: A set of questions.
  2. Contexts: Retrieved contexts corresponding to each question. This is a list[list] since each question can retrieve multiple text chunks.
  3. Answer: Generated answer corresponding to each question.
  4. الحقائق الأساسية: الحقائق الأساسية المقابلة لكل سؤال.هذا هو "str" الذي يتوافق مع الإجابة المتوقعة لكل سؤال.

لم يتم استخدام context المرجعي في الوقت الحالي

        metrics=[
# 这几个都不需要原始的context
context_precision,
answer_relevancy,
faithfulness,
context_recall,
],

استخدام BGE

from ragas.llama_index import evaluate

flag_model = HuggingFaceEmbeddings(model_name="BAAI/bge-small-en-v1.5")
query_engine2 = build_query_engine(flag_model)
result = evaluate(query_engine2, metrics, test_questions, test_answers)

التكامل LangSmith

تحديد المتغيرات البيئية

export LANGCHAIN_TRACING_V2=true
export LANGCHAIN_ENDPOINT=https://api.smith.langchain.com
export LANGCHAIN_API_KEY=<your-api-key>
export LANGCHAIN_PROJECT=<your-project> # if not specified, defaults to "default"

إنشاء جهاز تعقب

# langsmith
from langchain.callbacks.tracers import LangChainTracer

tracer = LangChainTracer(project_name="callback-experiments")

استخدامها في التقييم

from datasets import load_dataset
from ragas.metrics import context_precision
from ragas import evaluate

dataset = load_dataset("explodinggradients/amnesty_qa","english")
evaluate(dataset["train"],metrics=[context_precision],callbacks=[tracer])

دمج LlamaIndex

官方文档有问题。