ssrjkk Pinecone
Managed vector database for semantic search and RAG.
Quick Start
from pinecone import Pinecone, ServerlessSpec
pc = Pinecone(api_key="YOUR_API_KEY")
pc.create_index(name="my-index", dimension=384, metric="cosine", spec=ServerlessSpec(cloud="aws", region="us-east-1"))
Indexing & Query
index = pc.Index("my-index")
index.upsert(vectors=[{"id": "doc1", "values": [0.1, 0.2], "metadata": {"text": "Document 1"}}])
results = index.query(vector=query_embedding, top_k=5, include_metadata=True)
When to Use
- Semantic search; RAG vector storage; Recommendation systems
Validation
- Index creation succeeds; 2. Upsert operations complete; 3. Query returns relevant results