Image Similarity Search Pipeline

from mixpeek import Mixpeek

client = Mixpeek(api_key="YOUR_API_KEY", namespace="image-search")

# 1. A bucket for the images, and a collection that embeds each one with SigLIP
bucket = client.buckets.create(
    bucket_name="product-images",
    bucket_schema={
        "properties": {
            "image": {
                "type": "image",
            },
        },
    },
)
collection = client.collections.create(
    collection_name="products",
    source={"type": "bucket", "bucket_ids": [bucket["bucket_id"]]},
    feature_extractor={
        "feature_extractor_name": "image_extractor",
        "version": "v1",
    },
)

# 2. Upload and process
client.buckets.upload(
    bucket["bucket_id"],
    blobs=[{"property": "image", "type": "image", "data": "s3://your-bucket/product-images/handbag-0142.jpg"}],
)
client.collections.trigger(collection["collection_id"])

# 3. A text query searched against the image embeddings
retriever = client.retrievers.create(
    retriever_name="image-search",
    collection_identifiers=["products"],
    input_schema={
        "query": {
            "type": "text",
            "required": True,
        },
    },
    stages=[
        {
            "stage_name": "search",
            "stage_id": "feature_search",
            "parameters": {
                "searches": [
                    {
                        "feature_uri": "mixpeek://image_extractor@v1/google_siglip_base_v1",
                        "query": {
                            "input_mode": "text",
                            "value": "{{INPUT.query}}",
                        },
                        "top_k": 20,
                    },
                ],
                "final_top_k": 20,
            },
        },
    ],
)

# 4. Search
results = client.retrievers.execute(
    retriever["retriever_id"],
    inputs={
    "query": "red leather handbag with gold hardware",
},
)

for doc in results["documents"]:
    print(doc["document_id"], doc["score"])