BYO Embeddings Vector Search

from openai import OpenAI

from mixpeek import Mixpeek

openai = OpenAI(api_key="YOUR_OPENAI_KEY")
client = Mixpeek(api_key="YOUR_API_KEY")
NAMESPACE = "byo-docs"


def embed(text):
    return openai.embeddings.create(model="text-embedding-3-small", input=text).data[0].embedding


# 1. A standalone namespace for vectors you bring. text-embedding-3-small returns 1536 numbers.
client.namespaces.create(
    namespace_id=NAMESPACE,
    mode="standalone",
    vector_configs=[{"name": "dense", "dimension": 1536, "metric": "cosine"}],
)

# 2. Upsert each document with its named vector and a payload
documents = [
    {"id": "doc-1", "text": "How to deploy a Kubernetes cluster", "category": "devops"},
    {"id": "doc-2", "text": "Introduction to neural network architectures", "category": "ml"},
    {"id": "doc-3", "text": "Building REST APIs with FastAPI", "category": "backend"},
]
client.namespaces.documents.upsert(
    namespace_id=NAMESPACE,
    documents=[
        {"document_id": d["id"], "vectors": {"dense": embed(d["text"])}, "payload": {"text": d["text"], "category": d["category"]}}
        for d in documents
    ],
)

# 3. Search with a query vector. feature_uri is the vector name you upserted under.
results = client.search(
    namespace_id=NAMESPACE,
    queries=[{
        "feature_uri": "dense",
        "query": {"input_mode": "vector", "value": embed("how to set up container orchestration")},
        "top_k": 5,
    }],
)
for doc in results["documents"]:
    print(round(doc["score"], 3), doc.get("category"), doc.get("text"))