RAG-Chatbot mit Pinecone und OpenAI

Vektordatenbanken für Einbettungen mit Pinecone

James Chapman

Curriculum Manager, DataCamp

Retrieval Augmented Generation (RAG)

Benutzeranfrage einbetten
Ähnliche Dokumente abrufen
Dokumente zum Prompt hinzufügen

Retrieval Augmented Generation

Pinecone und OpenAI initialisieren

from openai import OpenAI
from pinecone import Pinecone
import pandas as pd
from uuid import uuid4


client = OpenAI(api_key="OPENAI_API_KEY")
pc = Pinecone(api_key="PINECONE_API_KEY")


index = pc.Index("semantic-search-datacamp")

YouTube-Transkripte

youtube_df = pd.read_csv('youtube_rag_data.csv')

| id | blob | channel_id | end | published | start | text | title | url |
|----|------|------------|-----|-----------|-------|------|-------|-----|
|int | dict | str        | int | datetime  | int   | str  | str   | str |

Dokumente ingestieren

batch_limit = 100


for batch in np.array_split(youtube_df, len(youtube_df) / batch_limit):

    metadatas = [{"text_id": row['id'], "text": row['text'],
      "title": row['title'], "url": row['url'], 
      "published": row['published']} for  _, row in batch.iterrows()]

    texts = batch['text'].tolist()

    ids = [str(uuid4()) for _ in range(len(texts))]


    response = client.embeddings.create(input=texts, model="text-embedding-3-small")
    embeds = [np.array(x.embedding) for x in response.data]


    index.upsert(vectors=zip(ids, embeds, metadatas), namespace='youtube_rag_dataset')

Retrieval-Funktion

def retrieve(query, top_k, namespace, emb_model):

    query_response = client.embeddings.create(input=query, model=emb_model)
    query_emb = query_response.data[0].embedding


    retrieved_docs = []
    sources = []
    docs = index.query(vector=query_emb, top_k=top_k,
                       namespace='youtube_rag_dataset', include_metadata=True)


    for doc in docs['matches']:
        retrieved_docs.append(doc['metadata']['text'])
        sources.append((doc['metadata']['title'], doc['metadata']['url']))

    return retrieved_docs, sources

Retrieval-Ausgabe

query = "How to build next-level Q&A with OpenAI"
documents, sources = retrieve(query, top_k=3, namespace='youtube_rag_dataset',
                              emb_model="text-embedding-3-small")

Document: Für Open-Domain-Fragenbeantwortung. Wir starten damit, ...
Source: How to build a Q&A AI in Python [...], https://youtu.be/w1dMEWm7jBc

Document: Hier haben wir Google und können Google Fragen stellen ...
Source: How to build next-level Q&A with OpenAI, https://youtu.be/coaaSxys5so

Document: Wir brauchen eine Vektordatenbank, um die Qualität von Q&A-Systemen zu verbessern ...
Source: How to Build Custom Q&A Transfo [...], https://youtu.be/ZIRmXKHp0-c

Prompt-Funktion mit Kontext

def prompt_with_context_builder(query, docs):
    delim = '\n\n---\n\n'
    prompt_start = 'Answer the question based on the context below.\n\nContext:\n'
    prompt_end = f'\n\nQuestion: {query}\nAnswer:'

    prompt = prompt_start + delim.join(docs) + prompt_end
    return prompt

Ausgabe: Prompt mit Kontext

query = "How to build next-level Q&A with OpenAI"
context_prompt = prompt_with_context_builder(query, documents)

Die Ausgabe der Funktion prompt_with_context_builder

Frage-Antwort-Funktion

def question_answering(prompt, sources, chat_model):

    sys_prompt = "You are a helpful assistant that always answers questions."

    res = client.chat.completions.create(
        model=chat_model,
        messages=[{"role": "system", "content": sys_prompt},
            {"role": "user", "content": prompt}
        ], temperature=0)

    answer = res.choices[0].message.content.strip()
    answer += "\n\nSources:"
    for source in sources:
        answer += "\n" + source[0] + ": " + source[1]
    return answer

Frage-Antwort-Ausgabe

query = "How to build next-level Q&A with OpenAI"
answer = question_answering(prompt_with_context, sources,
                            chat_model='gpt-4o-mini')

Die Ausgabe der Funktion question_answering

Alles zusammenführen

query = "How to build next-level Q&A with OpenAI"
documents, sources = retrieve(query, top_k=3, 
                              namespace='youtube_rag_dataset', 
                              emb_model="text-embedding-3-small")
prompt_with_context = prompt_with_context_builder(query, documents)
answer = question_answering(prompt_with_context, sources,
                            chat_model='gpt-4o-mini')

Lass uns üben!

Vektordatenbanken für Einbettungen mit Pinecone