Vector Databases: Pinecone, Weaviate & pgvector · Leçon

Gérer les espaces de noms

Comprenez comment segmenter vos données au sein d’un même index Pinecone à l’aide d’espaces de noms, pour une meilleure organisation.

Gérer les espaces de noms est une leçon Vector Databases: Pinecone, Weaviate & pgvector gratuite sur CoddyKit. Ceci est la leçon 2 sur 4. Tu peux lire la leçon complète ci-dessous gratuitement — puis la pratiquer en direct dans le navigateur avec un éditeur de code intégré et un tuteur IA 24/7. Elle fait partie du parcours d'apprentissage Vector Databases: Pinecone, Weaviate & pgvector, et ta progression se synchronise sur le web et l'application CoddyKit. Le cours Vector Databases: Pinecone, Weaviate & pgvector comprend 4 leçons au total.

Certaines parties de cette leçon n'ont pas encore été traduites et s'affichent en anglais.

What are Pinecone Namespaces?

Welcome to managing namespaces in Pinecone! A namespace is like a logical partition within a single Pinecone index.

Think of it as a folder inside a main directory. It allows you to segment your data without needing to create entirely separate indexes.

Why Use Namespaces?

Namespaces are incredibly useful for organizing your vector data, especially in complex applications:

Multi-tenancy: Isolate data for different users or customers within one index.
A/B Testing: Store different versions of embeddings for experiments.
Data Isolation: Keep distinct datasets separate while leveraging the same index infrastructure.
Categorization: Group vectors by topic, source, or type.

The Default Namespace

Every Pinecone index has a default namespace. If you don't explicitly specify a namespace when upserting or querying, your operations will interact with this default namespace.

It's often represented by an empty string "" or simply by omitting the namespace parameter.

Setting Up for Namespaces

Before we work with namespaces, ensure your Pinecone client is initialized. We'll assume you have your API key and environment configured, typically via environment variables.

Here's a basic setup for demonstration:

from pinecone import Pinecone
import os

def main():
    # Replace with your actual API key and environment from environment variables
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    print("Pinecone client initialized.")

if __name__ == "__main__":
    main()

Upserting Data into a Namespace

To add vectors to a specific namespace, you simply include the namespace parameter in your index.upsert() call.

This example upserts a few vectors into a namespace called "product-recommendations".

from pinecone import Pinecone, Index
import os

def main():
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    index_name = "my-product-index" # Ensure this index exists

    if index_name not in pc.list_indexes():
        print(f"Index '{index_name}' does not exist. Please create it first.")
        return

    index = pc.Index(index_name)

    vectors_to_upsert = [
        {"id": "item-1", "values": [0.1, 0.2, 0.3]}, # 3 dimensions for example
        {"id": "item-2", "values": [0.4, 0.5, 0.6]}
    ]

    # Upsert into a specific namespace
    index.upsert(vectors=vectors_to_upsert, namespace="product-recommendations")
    print("Vectors upserted into 'product-recommendations' namespace.")

if __name__ == "__main__":
    main()

Querying a Specific Namespace

When you want to retrieve vectors from a particular namespace, you also specify the namespace parameter in your index.query() call.

This ensures your search is confined to that segment of your index.

from pinecone import Pinecone, Index
import os

def main():
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    index_name = "my-product-index"

    if index_name not in pc.list_indexes():
        print(f"Index '{index_name}' does not exist. Please create it first.")
        return

    index = pc.Index(index_name)

    query_vector = [0.15, 0.25, 0.35] # Example query vector

    # Query within the 'product-recommendations' namespace
    results = index.query(
        vector=query_vector,
        top_k=2,
        namespace="product-recommendations",
        include_values=False
    )
    print("Query results from 'product-recommendations' namespace:")
    for match in results.matches:
        print(f"ID: {match.id}, Score: {match.score}")

if __name__ == "__main__":
    main()

Understanding Query Scope

It's crucial to remember that queries in Pinecone are namespace-specific by default. If you query without specifying a namespace, it will only search the default namespace.

There's no direct way to query across all namespaces with a single API call; you'd need to query each namespace individually if you wanted to combine results.

Listing Namespaces in an Index

To see which namespaces currently exist in your index and their statistics, you can use the index.describe_index_stats() method.

This returns a summary including a map of namespaces and their vector counts.

from pinecone import Pinecone, Index
import os

def main():
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    index_name = "my-product-index"

    if index_name not in pc.list_indexes():
        print(f"Index '{index_name}' does not exist. Please create it first.")
        return

    index = pc.Index(index_name)

    # Get index statistics
    index_stats = index.describe_index_stats()

    print("Namespaces in index:")
    for ns, stats in index_stats.namespaces.items():
        print(f"- Namespace: '{ns}', Vector Count: {stats.vector_count}")

if __name__ == "__main__":
    main()

Deleting Vectors from a Namespace

You can delete specific vectors from a namespace by providing their IDs along with the namespace parameter to index.delete().

This example deletes a specific item from the "product-recommendations" namespace.

from pinecone import Pinecone, Index
import os

def main():
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    index_name = "my-product-index"

    if index_name not in pc.list_indexes():
        print(f"Index '{index_name}' does not exist. Please create it first.")
        return

    index = pc.Index(index_name)

    # Delete a specific vector by ID within a namespace
    index.delete(ids=["item-1"], namespace="product-recommendations")
    print("Vector 'item-1' deleted from 'product-recommendations' namespace.")

if __name__ == "__main__":
    main()

Deleting an Entire Namespace

To remove all vectors within a specific namespace, effectively deleting the namespace itself, you can use index.delete() with the namespace parameter and set delete_all=True.

Be careful: This action is irreversible and deletes all data in that namespace!

from pinecone import Pinecone, Index
import os

def main():
    api_key = os.environ.get("PINECONE_API_KEY", "YOUR_API_KEY")
    environment = os.environ.get("PINECONE_ENVIRONMENT", "YOUR_ENVIRONMENT")

    if api_key == "YOUR_API_KEY" or environment == "YOUR_ENVIRONMENT":
        print("Please set PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars.")
        return

    pc = Pinecone(api_key=api_key, environment=environment)
    index_name = "my-product-index"

    if index_name not in pc.list_indexes():
        print(f"Index '{index_name}' does not exist. Please create it first.")
        return

    index = pc.Index(index_name)

    # Delete all vectors within a specific namespace
    # This effectively removes the 'test-namespace' and all its contents
    index.delete(namespace="test-namespace", delete_all=True)
    print("All vectors in 'test-namespace' deleted. Namespace effectively removed.")

if __name__ == "__main__":
    main()

Namespace Knowledge Check

You have an index named "my-app-index". You upsert vectors into "user-1" and "user-2" namespaces. If you then call index.query(vector=my_vec, top_k=5) without specifying a namespace, what will happen?

Recap: Organize with Namespaces

Great job! You've learned about the power of Pinecone namespaces.

Namespaces help segment data within a single index.
They're perfect for multi-tenancy and A/B testing.
Operations (upsert, query, delete) are namespace-specific.
The default namespace is used if none is specified.
You can list existing namespaces and delete entire namespaces.

Using namespaces effectively keeps your vector data organized and manageable!

Questions Fréquemment Posées

La leçon « Gérer les espaces de noms » est-elle gratuite ?

Oui — le texte complet de « Gérer les espaces de noms » est gratuit à lire ici sur le web. Pour la pratiquer de manière interactive (un éditeur de code intégré et un tuteur IA 24/7) et déverrouiller le reste du cours Vector Databases: Pinecone, Weaviate & pgvector, passe à CoddyKit PRO. Le cours Vector Databases: Pinecone, Weaviate & pgvector comprend 4 leçons au total.

Qu'est-ce que j'apprendrai dans « Gérer les espaces de noms » ?

Comprenez comment segmenter vos données au sein d’un même index Pinecone à l’aide d’espaces de noms, pour une meilleure organisation. Tu pratiques Vector Databases: Pinecone, Weaviate & pgvector avec du code pratique que tu exécutes directement dans le navigateur, et un tuteur IA 24/7 répond à tes questions au fur et à mesure que tu avances dans la leçon.

Dois-je avoir de l'expérience pour commencer Vector Databases: Pinecone, Weaviate & pgvector ?

Aucune expérience préalable n'est requise. Vector Databases: Pinecone, Weaviate & pgvector sur CoddyKit est structuré pour les débutants jusqu'aux apprenants avancés, donc tu peux commencer ici ou depuis le début et avancer à ton rythme. Ceci est la leçon 2 sur 4.

Combien de temps prend la leçon « Gérer les espaces de noms » ?

La plupart des leçons CoddyKit prennent environ 5–10 minutes. Chacune est courte et interactive, tu progresses régulièrement et tu repiques exactement où tu t'es arrêté sur le web et l'app.

Peux-tu écrire et exécuter du code dans cette leçon Vector Databases: Pinecone, Weaviate & pgvector ?

Oui. Chaque leçon Vector Databases: Pinecone, Weaviate & pgvector inclut un éditeur de code intégré, tu écris et exécutes du vrai code directement dans ton navigateur et tu reçois des retours IA instantanés — aucune configuration locale requise.

Toutes les leçons de ce cours

Filtrer avec les métadonnées
Gérer les espaces de noms
Mises à jour et suppressions en temps réel
Recherche hybride avec vecteurs creux et denses

← Retour à Vector Databases: Pinecone, Weaviate & pgvector