Del via

Evaluering af grundlæggende RAG-ydeevne

Dette selvstudium viser, hvordan du bruger Fabric til at evaluere ydeevnen for RAG-programmer. Evalueringen fokuserer på to primære RAG-komponenter: retrieveren (Azure AI Search) og svargeneratoren (en LLM, der bruger brugerens forespørgsel, hentede kontekst og en prompt til at generere et svar). Her er de vigtigste trin:

  1. Konfigurer Azure OpenAI- og Azure AI Search-tjenester
  2. Indlæs data fra CMU's QA-datasæt med Wikipedia-artikler for at opbygge et benchmark
  3. Kør en røgtest med én forespørgsel for at bekræfte, at RAG-systemet fungerer fra ende til anden
  4. Definer deterministiske og AI-assisterede målinger til evaluering
  5. Check-in 1: Evaluer retriever-ydeevne ved hjælp af top-N-nøjagtighed
  6. Check-in 2: Evaluer responsgeneratorens ydeevne ved hjælp af jordforbindelse, relevans og lighedsmålinger
  7. Visualiser og gem evalueringsresultater i OneLake til fremtidig reference og løbende evaluering

Forudsætninger

Før du starter dette selvstudium, skal du fuldføre den trinvise vejledning til udvidelse af bygningshentning i stof.

Du skal bruge disse tjenester for at køre notesbogen:

I den forrige vejledning uploadede du data til dit søhus og byggede et dokumentindeks, der bruges af RAG-systemet. Brug indekset i denne øvelse til at lære kerneteknikker til at evaluere RAG-ydeevne og identificere potentielle problemer. Hvis du ikke har oprettet et indeks eller fjernet det, skal du følge hurtigstartsvejledningen for at fuldføre forudsætningen.

Diagram, der viser flowet i en brugersamtale gennem RAG-systemet.

Definer slutpunkter og påkrævede nøgler. Importer nødvendige biblioteker og funktioner. Instantier klienter til Azure OpenAI og Azure AI Search. Definer en funktionsombryder med en prompt om at forespørge på RAG-systemet.

# Enter your Azure OpenAI service values
aoai_endpoint = "https://<your-resource-name>.openai.azure.com" # TODO: Provide the Azure OpenAI resource endpoint (replace <your-resource-name>)
aoai_key = "" # TODO: Fill in your API key from Azure OpenAI 
aoai_deployment_name_embeddings = "text-embedding-ada-002"
aoai_model_name_query = "gpt-4-32k"  
aoai_model_name_metrics = "gpt-4-32k"
aoai_api_version = "2024-02-01"

# Setup key accesses to Azure AI Search
aisearch_index_name = "" # TODO: Create a new index name: must only contain lowercase, numbers, and dashes
aisearch_api_key = "" # TODO: Fill in your API key from Azure AI Search
aisearch_endpoint = "https://.search.windows.net" # TODO: Provide the url endpoint for your created Azure AI Search 

import warnings
warnings.filterwarnings("ignore", category=DeprecationWarning) 

import os, requests, json

from datetime import datetime, timedelta
from azure.core.credentials import AzureKeyCredential
from azure.search.documents import SearchClient

from pyspark.sql import functions as F
from pyspark.sql.functions import to_timestamp, current_timestamp, concat, col, split, explode, udf, monotonically_increasing_id, when, rand, coalesce, lit, input_file_name, regexp_extract, concat_ws, length, ceil
from pyspark.sql.types import StructType, StructField, StringType, IntegerType, TimestampType, ArrayType, FloatType
from pyspark.sql import Row
import pandas as pd
from azure.search.documents.indexes import SearchIndexClient
from azure.search.documents.models import (
    VectorizedQuery,
)
from azure.search.documents.indexes.models import (  
    SearchIndex,  
    SearchField,  
    SearchFieldDataType,  
    SimpleField,  
    SearchableField,   
    SemanticConfiguration,  
    SemanticPrioritizedFields,
    SemanticField,  
    SemanticSearch,
    VectorSearch,  
    HnswAlgorithmConfiguration,
    HnswParameters,  
    VectorSearchProfile,
    VectorSearchAlgorithmKind,
    VectorSearchAlgorithmMetric,
)

import openai 
from openai import AzureOpenAI
import uuid
import matplotlib.pyplot as plt
from synapse.ml.featurize.text import PageSplitter
import ipywidgets as widgets  
from IPython.display import display as w_display

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 6, Finished, Available, Finished)

# Configure access to OpenAI endpoint
openai.api_type = "azure"
openai.api_key = aoai_key
openai.api_base = aoai_endpoint
openai.api_version = aoai_api_version

# Create client for accessing embedding endpoint
embed_client = AzureOpenAI(
    api_version=aoai_api_version,
    azure_endpoint=aoai_endpoint,
    api_key=aoai_key,
)

# Create client for accessing chat endpoint
chat_client = AzureOpenAI(
    azure_endpoint=aoai_endpoint,
    api_key=aoai_key,
    api_version=aoai_api_version,
)

# Configure access to Azure AI Search
search_client = SearchClient(
    aisearch_endpoint,
    aisearch_index_name,
    credential=AzureKeyCredential(aisearch_api_key)
)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 7, Finished, Available, Finished)

Følgende funktioner implementerer de to vigtigste RAG-komponenter - retriever (get_context_source) og responsgenerator (get_answer). Koden ligner den forrige tutorial. Parameteren topN giver dig mulighed for at angive, hvor mange relevante ressourcer der skal hentes (dette selvstudium bruger 3, men den optimale værdi kan variere efter datasæt):

# Implement retriever
def get_context_source(question, topN=3):
    """
    Retrieves contextual information and sources related to a given question using embeddings and a vector search.  
    Parameters:  
    question (str): The question for which the context and sources are to be retrieved.  
    topN (int, optional): The number of top results to retrieve. Default is 3.  
      
    Returns:  
    List: A list containing two elements:  
        1. A string with the concatenated retrieved context.  
        2. A list of retrieved source paths.  
    """
    embed_client = openai.AzureOpenAI(
        api_version=aoai_api_version,
        azure_endpoint=aoai_endpoint,
        api_key=aoai_key,
    )

    query_embedding = embed_client.embeddings.create(input=question, model=aoai_deployment_name_embeddings).data[0].embedding

    vector_query = VectorizedQuery(vector=query_embedding, k_nearest_neighbors=topN, fields="Embedding")

    results = search_client.search(   
        vector_queries=[vector_query],
        top=topN,
    )

    retrieved_context = ""
    retrieved_sources = []
    for result in results:
        retrieved_context += result['ExtractedPath'] + "\n" + result['Chunk'] + "\n\n"
        retrieved_sources.append(result['ExtractedPath'])

    return [retrieved_context, retrieved_sources]

# Implement response generator
def get_answer(question, context):
    """  
    Generates a response to a given question using provided context and an Azure OpenAI model.  
    
    Parameters:  
        question (str): The question that needs to be answered.  
        context (str): The contextual information related to the question that will help generate a relevant response.  
    
    Returns:  
        str: The response generated by the Azure OpenAI model based on the provided question and context.  
    """
    messages = [
        {
            "role": "system",
            "content": "You are a chat assistant. Use provided text to ground your response. Give a one-word answer when possible ('yes'/'no' is OK where appropriate, no details). Unnecessary words incur a $500 penalty."
        }
    ]

    messages.append(
        {
            "role": "user", 
            "content": question + "\n" + context,
        },
    )

    chat_client = openai.AzureOpenAI(
        azure_endpoint=aoai_endpoint,
        api_key=aoai_key,
        api_version=aoai_api_version,
    )

    chat_completion = chat_client.chat.completions.create(
        model=aoai_model_name_query,
        messages=messages,
    )

    return chat_completion.choices[0].message.content

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 8, Finished, Available, Finished)

Datasæt

Version 1.2 af Carnegie Mellon University Question-Answer-datasættet er et korpus af Wikipedia-artikler med faktuelle spørgsmål og svar skrevet manuelt. Den hostes i Azure Blob Storage under GFDL. Datasættet bruger én tabel med disse felter:

  • ArticleTitle: Navnet på den Wikipedia-artikel, som spørgsmålene og svarene kommer fra
  • Question: Manuelt skrevet spørgsmål om artiklen
  • Answer: Manuelt skrevet svar baseret på artiklen
  • DifficultyFromQuestioner: Bedømmelse af sværhedsgrad, som forfatteren tildeler
  • DifficultyFromAnswerer: Sværhedsgrad, som evaluatoren tildeler; kan afvige fra DifficultyFromQuestioner
  • ExtractedPath: Sti til den oprindelige artikel (en artikel kan have flere spørgsmål-svar-par)
  • text: Renset Wikipedia-artikeltekst

Download LICENSE-S08- og LICENSE-S09 filerne fra samme placering for at få licensoplysninger.

Historie og citat

Brug dette citat til datasættet:

CMU Question/Answer Dataset, Release 1.2
August 23, 2013
Noah A. Smith, Michael Heilman, and Rebecca Hwa
Question Generation as a Competitive Undergraduate Course Project
In Proceedings of the NSF Workshop on the Question Generation Shared Task and Evaluation Challenge, Arlington, VA, September 2008. 
Available at http://www.cs.cmu.edu/~nasmith/papers/smith+heilman+hwa.nsf08.pdf.
Original dataset acknowledgments:
This research project was supported by NSF IIS-0713265 (to Smith), an NSF Graduate Research Fellowship (to Heilman), NSF IIS-0712810 and IIS-0745914 (to Hwa), and Institute of Education Sciences, U.S. Department of Education R305B040063 (to Carnegie Mellon).
cmu-qa-08-09 (modified version)
June 12, 2024
Amir Jafari, Alexandra Savelieva, Brice Chung, Hossein Khadivi Heris, Journey McDowell
This release uses the GNU Free Documentation License (GFDL) (http://www.gnu.org/licenses/fdl.html).
The GNU license applies to all copies of the dataset.

Opret benchmark

Importer benchmarket. Til denne demo skal du bruge et undersæt af spørgsmål fra S08/set1 og S08/set2 buckets. For at beholde et spørgsmål pr. artikel, skal du anvende df.dropDuplicates(["ExtractedPath"]). Slip duplikerede spørgsmål. Kurateringsprocessen tilføjer sværhedsgrader; Dette eksempel begrænser dem til medium.

df = spark.sql("SELECT * FROM data_load_tests.cmu_qa")

# Filter the DataFrame to include the specified paths
df = df.filter((col("ExtractedPath").like("S08/data/set1/%")) | (col("ExtractedPath").like("S08/data/set2/%")))

# Keep only medium-difficulty questions.
df = df.filter(col("DifficultyFromQuestioner") == "medium")


# Drop duplicate questions and source paths.
df = df.dropDuplicates(["Question"])
df = df.dropDuplicates(["ExtractedPath"])

num_rows = df.count()
num_columns = len(df.columns)
print(f"Number of rows: {num_rows}, Number of columns: {num_columns}")

# Persist the DataFrame
df.persist()
display(df)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 9, Finished, Available, Finished)Number of rows: 20, Number of columns: 7SynapseWidget(Synapse.DataFrame, 47aff8cb-72f8-4a36-885c-f4f3bb830a91)

Resultatet er en DataFrame med 20 rækker - demo-benchmarket. Nøglefelter er Question, Answer (menneskeligt kurateret jordsandhedssvar) og ExtractedPath (kildedokumentet). Juster filtrene for at inkludere andre spørgsmål og variere sværhedsgraden for at få et mere realistisk eksempel. Prøv det.

Kør en simpel end-to-end-test

Start med en end-to-end røgtest af hentningsforstærket generering (RAG).

question = "How many suborders are turtles divided into?"
retrieved_context, retrieved_sources = get_context_source(question)
answer = get_answer(question, retrieved_context)
print(answer)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 10, Finished, Available, Finished)Three

Denne røgtest hjælper dig med at finde problemer i RAG-implementeringen, f.eks. forkerte legitimationsoplysninger, et manglende eller tomt vektorindeks eller inkompatible funktionsgrænseflader. Hvis testen mislykkes, skal du kontrollere, om der er problemer. Forventet output: Three. Hvis røgtesten består, skal du gå til næste afsnit for at evaluere RAG yderligere.

Etablere målinger

Definer en deterministisk metrik for at evaluere retrieveren. Den er inspireret af søgemaskiner. Den kontrollerer, om listen over hentede kilder indeholder den grundlæggende sandhedskilde Denne metrikværdi er en top-N-nøjagtighedsscore, fordi topN parameteren angiver antallet af hentede kilder.

def get_retrieval_score(target_source, retrieved_sources):
    if target_source in retrieved_sources: 
        return 1
    else: 
        return 0

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 11, Finished, Available, Finished)

Ifølge benchmarket er svaret indeholdt i kilden med ID "S08/data/set1/a9". Test af funktionen på eksemplet, vi kørte ovenfor, returnerer 1som forventet, fordi den var i de tre øverste relevante tekstsegmenter.

print("Retrieved sources:", retrieved_sources)
get_retrieval_score("S08/data/set1/a9", retrieved_sources)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 12, Finished, Available, Finished)Retrieved sources: ['S08/data/set1/a9', 'S08/data/set1/a9', 'S08/data/set1/a5']1

I dette afsnit defineres AI-assisterede målepunkter. Promptskabelonen indeholder et par eksempler på input (KONTEKST og SVAR) og foreslået output - også kendt som en model med få skud. Det er den samme prompt, der bruges i Azure AI Studio. Få mere at vide i Indbyggede evalueringsmålepunkter. Denne demo bruger og-målingerne groundednessrelevance - disse er normalt de mest nyttige og pålidelige til evaluering af GPT-modeller. Andre målinger kan være nyttige, men giver mindre intuition - for eksempel behøver svarene ikke at være ens for at være korrekte, så similarity scorer kan være vildledende. Skalaen for alle målepunkter er 1 til 5. Højere er bedre. Jordnærhed kræver kun to input (kontekst og genereret svar), mens de to andre målinger også bruger grundsandhed til evaluering.

def get_groundedness_metric(context, answer):
    """Get the groundedness score from the LLM using the context and answer."""

    groundedness_prompt_template = """
    You are presented with a CONTEXT and an ANSWER about that CONTEXT. Decide whether the ANSWER is entailed by the CONTEXT by choosing one of the following ratings:
    1. 5: The ANSWER follows logically from the information contained in the CONTEXT.
    2. 1: The ANSWER is logically false from the information contained in the CONTEXT.
    3. an integer score between 1 and 5 and if such integer score does not exist, use 1: It is not possible to determine whether the ANSWER is true or false without further information. Read the passage of information thoroughly and select the correct answer from the three answer labels. Read the CONTEXT thoroughly to ensure you know what the CONTEXT entails. Note the ANSWER is generated by a computer system, it can contain certain symbols, which should not be a negative factor in the evaluation.
    Independent Examples:
    ## Example Task #1 Input:
    "CONTEXT": "Some are reported as not having been wanted at all.", "QUESTION": "", "ANSWER": "All are reported as being completely and fully wanted."
    ## Example Task #1 Output:
    1
    ## Example Task #2 Input:
    "CONTEXT": "Ten new television shows appeared during the month of September. Five of the shows were sitcoms, three were hourlong dramas, and two were news-magazine shows. By January, only seven of these new shows were still on the air. Five of the shows that remained were sitcoms.", "QUESTION": "", "ANSWER": "At least one of the shows that were cancelled was an hourlong drama."
    ## Example Task #2 Output:
    5
    ## Example Task #3 Input:
    "CONTEXT": "In Quebec, an allophone is a resident, usually an immigrant, whose mother tongue or home language is neither French nor English.", "QUESTION": "", "ANSWER": "In Quebec, an allophone is a resident, usually an immigrant, whose mother tongue or home language is not French."
    5
    ## Example Task #4 Input:
    "CONTEXT": "Some are reported as not having been wanted at all.", "QUESTION": "", "ANSWER": "All are reported as being completely and fully wanted."
    ## Example Task #4 Output:
    1
    ## Actual Task Input:
    "CONTEXT": {context}, "QUESTION": "", "ANSWER": {answer}
    Reminder: The return values for each task should be correctly formatted as an integer between 1 and 5. Do not repeat the context and question.  Don't explain the reasoning. The answer should include only a number: 1, 2, 3, 4, or 5.
    Actual Task Output:
    """

    metric_client = openai.AzureOpenAI(
        api_version=aoai_api_version,
        azure_endpoint=aoai_endpoint,
        api_key=aoai_key,
    )

    messages = [
        {
            "role": "system",
            "content": "You are an AI assistant. You will be given the definition of an evaluation metric for assessing the quality of an answer in a question-answering task. Your job is to compute an accurate evaluation score using the provided evaluation metric."
        }, 
        {
            "role": "user",
            "content": groundedness_prompt_template.format(context=context, answer=answer)
        }
    ]

    metric_completion = metric_client.chat.completions.create(
        model=aoai_model_name_metrics,
        messages=messages,
        temperature=0,
    )

    return metric_completion.choices[0].message.content

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 13, Finished, Available, Finished)

def get_relevance_metric(context, question, answer):    
    relevance_prompt_template = """
    Relevance measures how well the answer addresses the main aspects of the question, based on the context. Consider whether all and only the important aspects are contained in the answer when evaluating relevance. Given the context and question, score the relevance of the answer between one to five stars using the following rating scale:
    One star: the answer completely lacks relevance
    Two stars: the answer mostly lacks relevance
    Three stars: the answer is partially relevant
    Four stars: the answer is mostly relevant
    Five stars: the answer has perfect relevance

    This rating value should always be an integer between 1 and 5. So the rating produced should be 1 or 2 or 3 or 4 or 5.

    context: Marie Curie was a Polish-born physicist and chemist who pioneered research on radioactivity and was the first woman to win a Nobel Prize.
    question: What field did Marie Curie excel in?
    answer: Marie Curie was a renowned painter who focused mainly on impressionist styles and techniques.
    stars: 1

    context: The Beatles were an English rock band formed in Liverpool in 1960, and they are widely regarded as the most influential music band in history.
    question: Where were The Beatles formed?
    answer: The band The Beatles began their journey in London, England, and they changed the history of music.
    stars: 2

    context: The recent Mars rover, Perseverance, was launched in 2020 with the main goal of searching for signs of ancient life on Mars. The rover also carries an experiment called MOXIE, which aims to generate oxygen from the Martian atmosphere.
    question: What are the main goals of Perseverance Mars rover mission?
    answer: The Perseverance Mars rover mission focuses on searching for signs of ancient life on Mars.
    stars: 3

    context: The Mediterranean diet is a commonly recommended dietary plan that emphasizes fruits, vegetables, whole grains, legumes, lean proteins, and healthy fats. Studies have shown that it offers numerous health benefits, including a reduced risk of heart disease and improved cognitive health.
    question: What are the main components of the Mediterranean diet?
    answer: The Mediterranean diet primarily consists of fruits, vegetables, whole grains, and legumes.
    stars: 4

    context: The Queen's Royal Castle is a well-known tourist attraction in the United Kingdom. It spans over 500 acres and contains extensive gardens and parks. The castle was built in the 15th century and has been home to generations of royalty.
    question: What are the main attractions of the Queen's Royal Castle?
    answer: The main attractions of the Queen's Royal Castle are its expansive 500-acre grounds, extensive gardens, parks, and the historical castle itself, which dates back to the 15th century and has housed generations of royalty.
    stars: 5

    Don't explain the reasoning. The answer should include only a number: 1, 2, 3, 4, or 5.

    context: {context}
    question: {question}
    answer: {answer}
    stars:
    """

    metric_client = openai.AzureOpenAI(
        api_version=aoai_api_version,
        azure_endpoint=aoai_endpoint,
        api_key=aoai_key,
    )


    messages = [
        {
            "role": "system",
            "content": "You are an AI assistant. You are given the definition of an evaluation metric for assessing the quality of an answer in a question-answering task. Compute an accurate evaluation score using the provided evaluation metric."
        }, 
        {
            "role": "user",
            "content": relevance_prompt_template.format(context=context, question=question, answer=answer)
        }
    ]

    metric_completion = metric_client.chat.completions.create(
        model=aoai_model_name_metrics,
        messages=messages,
        temperature=0,
    )

    return metric_completion.choices[0].message.content

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 14, Finished, Available, Finished)

def get_similarity_metric(question, ground_truth, answer):
    similarity_prompt_template = """
    Equivalence, as a metric, measures the similarity between the predicted answer and the correct answer. If the information and content in the predicted answer is similar or equivalent to the correct answer, then the value of the Equivalence metric should be high, else it should be low. Given the question, correct answer, and predicted answer, determine the value of Equivalence metric using the following rating scale:
    One star: the predicted answer is not at all similar to the correct answer
    Two stars: the predicted answer is mostly not similar to the correct answer
    Three stars: the predicted answer is somewhat similar to the correct answer
    Four stars: the predicted answer is mostly similar to the correct answer
    Five stars: the predicted answer is completely similar to the correct answer

    This rating value should always be an integer between 1 and 5. So the rating produced should be 1 or 2 or 3 or 4 or 5.

    The examples below show the Equivalence score for a question, a correct answer, and a predicted answer.

    question: What is the role of ribosomes?
    correct answer: Ribosomes are cellular structures responsible for protein synthesis. They interpret the genetic information carried by messenger RNA (mRNA) and use it to assemble amino acids into proteins.
    predicted answer: Ribosomes participate in carbohydrate breakdown by removing nutrients from complex sugar molecules.
    stars: 1

    question: Why did the Titanic sink?
    correct answer: The Titanic sank after it struck an iceberg during its maiden voyage in 1912. The impact caused the ship's hull to breach, allowing water to flood into the vessel. The ship's design, lifeboat shortage, and lack of timely rescue efforts contributed to the tragic loss of life.
    predicted answer: The sinking of the Titanic was a result of a large iceberg collision. This caused the ship to take on water and eventually sink, leading to the death of many passengers due to a shortage of lifeboats and insufficient rescue attempts.
    stars: 2

    question: What causes seasons on Earth?
    correct answer: Seasons on Earth are caused by the tilt of the Earth's axis and its revolution around the Sun. As the Earth orbits the Sun, the tilt causes different parts of the planet to receive varying amounts of sunlight, resulting in changes in temperature and weather patterns.
    predicted answer: Seasons occur because of the Earth's rotation and its elliptical orbit around the Sun. The tilt of the Earth's axis causes regions to be subjected to different sunlight intensities, which leads to temperature fluctuations and alternating weather conditions.
    stars: 3

    question: How does photosynthesis work?
    correct answer: Photosynthesis is a process by which green plants and some other organisms convert light energy into chemical energy. This occurs as light is absorbed by chlorophyll molecules, and then carbon dioxide and water are converted into glucose and oxygen through a series of reactions.
    predicted answer: In photosynthesis, sunlight is transformed into nutrients by plants and certain microorganisms. Light is captured by chlorophyll molecules, followed by the conversion of carbon dioxide and water into sugar and oxygen through multiple reactions.
    stars: 4

    question: What are the health benefits of regular exercise?
    correct answer: Regular exercise can help maintain a healthy weight, increase muscle and bone strength, and reduce the risk of chronic diseases. It also promotes mental well-being by reducing stress and improving overall mood.
    predicted answer: Routine physical activity can contribute to maintaining ideal body weight, enhancing muscle and bone strength, and preventing chronic illnesses. In addition, it supports mental health by alleviating stress and augmenting general mood.
    stars: 5

    Don't explain the reasoning. The answer should include only a number: 1, 2, 3, 4, or 5.

    question: {question}
    correct answer:{ground_truth}
    predicted answer: {answer}
    stars:
    """
    
    metric_client = openai.AzureOpenAI(
        api_version=aoai_api_version,
        azure_endpoint=aoai_endpoint,
        api_key=aoai_key,
    )

    messages = [
        {
            "role": "system",
            "content": "You are an AI assistant. You will be given the definition of an evaluation metric for assessing the quality of an answer in a question-answering task. Your job is to compute an accurate evaluation score using the provided evaluation metric."
        }, 
        {
            "role": "user",
            "content": similarity_prompt_template.format(question=question, ground_truth=ground_truth, answer=answer)
        }
    ]

    metric_completion = metric_client.chat.completions.create(
        model=aoai_model_name_metrics,
        messages=messages,
        temperature=0,
    )

    return metric_completion.choices[0].message.content

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 15, Finished, Available, Finished)

Test relevansmetrikken:

get_relevance_metric(retrieved_context, question, answer)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 16, Finished, Available, Finished)'2'

En score på 5 betyder, at svaret er relevant. Følgende kode henter lighedsmetrikken:

get_similarity_metric(question, 'three', answer)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 17, Finished, Available, Finished)'5'

En score på 5 betyder, at svaret matcher det grundlæggende sandhedssvar, der er kurateret af en menneskelig ekspert. AI-assisterede metriske scorer kan svinge med det samme input. De er hurtigere end at bruge menneskelige dommere.

Evaluer RAG-ydeevne på benchmark Q&A

Opret funktionsombrydere, der skal køre i stor skala. Ombryd hver funktion, der slutter med _udf (forkortelse for user-defined function), så den overholder Spark-kravene (@udf(returnType=StructType([ ... ]))) og kører beregninger på store data hurtigere på tværs af klyngen.

# UDF wrappers for RAG components
@udf(returnType=StructType([  
    StructField("retrieved_context", StringType(), True),  
    StructField("retrieved_sources", ArrayType(StringType()), True)  
]))
def get_context_source_udf(question, topN=3):
    return get_context_source(question, topN)

@udf(returnType=StringType())
def get_answer_udf(question, context):
    return get_answer(question, context)


# UDF wrapper for retrieval score
@udf(returnType=StringType())
def get_retrieval_score_udf(target_source, retrieved_sources):
    return get_retrieval_score(target_source, retrieved_sources)


# UDF wrappers for AI-assisted metrics
@udf(returnType=StringType())
def get_groundedness_metric_udf(context, answer):
    return get_groundedness_metric(context, answer)

@udf(returnType=StringType())
def get_relevance_metric_udf(context, question, answer): 
    return get_relevance_metric(context, question, answer)

@udf(returnType=StringType())
def get_similarity_metric_udf(question, ground_truth, answer):
    return get_similarity_metric(question, ground_truth, answer)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 18, Finished, Available, Finished)

Check-in #1: Retrieverens ydeevne

Følgende kode opretter kolonnerne result og retrieval_score i benchmarket DataFrame. Disse kolonner inkluderer det RAG-genererede svar og en indikator for, om den kontekst, der gives til LLM, inkluderer den artikel, som spørgsmålet er baseret på.

df = df.withColumn("result", get_context_source_udf(df.Question)).select(df.columns+["result.*"])
df = df.withColumn('retrieval_score', get_retrieval_score_udf(df.ExtractedPath, df.retrieved_sources))
print("Aggregate Retrieval score: {:.2f}%".format((df.where(df["retrieval_score"] == 1).count() / df.count()) * 100))
display(df.select(["question", "retrieval_score",  "ExtractedPath", "retrieved_sources"]))

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 19, Finished, Available, Finished)Aggregate Retrieval score: 100.00%SynapseWidget(Synapse.DataFrame, 14efe386-836a-4765-bd88-b121f32c7cfc)

For alle spørgsmål henter retrieveren den korrekte kontekst, og i de fleste tilfælde er det den første post. Azure AI Search fungerer godt. Du undrer dig måske over, hvorfor konteksten i nogle tilfælde har to eller tre identiske værdier. Det er ikke en fejl - det betyder, at retrieveren henter fragmenter af den samme artikel, der ikke passer ind i en del under opdeling.

Check-in #2: ydeevne af responsgeneratoren

Send spørgsmålet og konteksten til LLM for at generere et svar. Gem den i generated_answer kolonnen i DataFrame:

df = df.withColumn('generated_answer', get_answer_udf(df.Question, df.retrieved_context))

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 20, Finished, Available, Finished)

Brug det genererede svar, grundsandhedssvar, spørgsmål og kontekst til at beregne målepunkter. Vis evalueringsresultater for hvert spørgsmål-svar-par:

df = df.withColumn('gpt_groundedness', get_groundedness_metric_udf(df.retrieved_context, df.generated_answer))
df = df.withColumn('gpt_relevance', get_relevance_metric_udf(df.retrieved_context, df.Question, df.generated_answer))
df = df.withColumn('gpt_similarity', get_similarity_metric_udf(df.Question, df.Answer, df.generated_answer))
display(df.select(["question", "answer", "generated_answer", "retrieval_score", "gpt_groundedness","gpt_relevance", "gpt_similarity"]))

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 21, Finished, Available, Finished)SynapseWidget(Synapse.DataFrame, 22b97d27-91e1-40f3-b888-3a3399de9d6b)

Hvad viser disse værdier? For at gøre dem lettere at fortolke skal du plotte histogrammer for jordforbindelse, relevans og lighed. LLM er mere detaljeret end menneskelige grundsandhedssvar, hvilket sænker lighedsmetrikken - omkring halvdelen af svarene er semantisk korrekte, men får fire stjerner som for det meste ens. De fleste værdier for alle tre målinger er 4 eller 5, hvilket tyder på, at RAG-ydeevnen er god. Der er et par outliers - for eksempel for spørgsmålet How many species of otter are there?, den genererede There are 13 species of ottermodel , hvilket er korrekt med høj relevans og lighed (5). Af en eller anden grund anså GPT den for at være dårligt funderet i den angivne kontekst og gav den en stjerne. I de tre andre tilfælde med mindst én AI-assisteret metrik af én stjerne, peger den lave score på et dårligt svar. LLM scorer lejlighedsvis forkert, men scorer normalt præcist.

# Convert Spark DataFrame to Pandas DataFrame
pandas_df = df.toPandas()

selected_columns = ['gpt_groundedness', 'gpt_relevance', 'gpt_similarity']
trimmed_df = pandas_df[selected_columns].astype(int)

# Define a function to plot histograms for the specified columns
def plot_histograms(dataframe, columns):
    # Set up the figure size and subplots
    plt.figure(figsize=(15, 5))
    for i, column in enumerate(columns, 1):
        plt.subplot(1, len(columns), i)
        # Filter the dataframe to only include rows with values 1, 2, 3, 4, 5
        filtered_df = dataframe[dataframe[column].isin([1, 2, 3, 4, 5])]
        filtered_df[column].hist(bins=range(1, 7), align='left', rwidth=0.8)
        plt.title(f'Histogram of {column}')
        plt.xlabel('Values')
        plt.ylabel('Frequency')
        plt.xticks(range(1, 6))
        plt.yticks(range(0, 20, 2))


# Call the function to plot histograms for the specified columns
plot_histograms(trimmed_df, selected_columns)

# Show the plots
plt.tight_layout()
plt.show()

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 24, Finished, Available, Finished)

Skærmbillede af histogrammer, der viser fordelingen af GPT-relevans og lighedsscore for de evaluerede spørgsmål.

Som et sidste trin skal du gemme benchmarkresultaterne i en tabel i dit søhus. Dette trin er valgfrit, men anbefales stærkt - det gør dine resultater mere nyttige. Når du ændrer noget i RAG'en (f.eks. ændrer prompten, opdaterer indekset eller bruger en anden GPT-model i svargeneratoren), skal du måle indvirkningen, kvantificere forbedringer og registrere regressioner.

# create name of experiment that is easy to refer to
friendly_name_of_experiment = "rag_tutorial_experiment_1"

# Note the current date and time  
time_of_experiment = current_timestamp()

# Generate a unique GUID for all rows
experiment_id = str(uuid.uuid4())

# Add two new columns to the Spark DataFrame
updated_df = df.withColumn("execution_time", time_of_experiment) \
                        .withColumn("experiment_id", lit(experiment_id)) \
                        .withColumn("experiment_friendly_name", lit(friendly_name_of_experiment))

# Store the updated DataFrame in the default lakehouse as a table named 'rag_experiment_runs'
table_name = "rag_experiment_run_demo1" 
updated_df.write.format("parquet").mode("append").saveAsTable(table_name)

Celle output:StatementMeta(, 21cb8cd3-7742-4c1f-8339-265e2846df1d, 28, Finished, Available, Finished)

Vend tilbage til eksperimentresultaterne når som helst for at gennemgå dem, sammenligne med nye eksperimenter og vælge den konfiguration, der fungerer bedst til produktionen.

Resumé

Brug AI-assisterede målepunkter og top-N-hentningshastighed til at opbygge din RAG-løsning (Retrieval-augmented Generation).

  • Last updated on