DP-800: Developing AI-Enabled Database Solutions

Question

HOTSPOT -

You have a Microsoft Fabric SQL database named SalesDB that includes a table called dbo.Products.

Modify SalesDB to satisfy these requirements:

Create a vector index on the appropriate column.
Use a supplied natural-language query vector.

How should you complete the Transact-SQL code? Select the appropriate options in the answer area.

Select

CREATE VECTOR INDEX idx_products_embedding ON dbo.Products (  )
WITH (METRIC = 'cosine', TYPE = 'DiskANN');

DECLARE @query_vector VECTOR(1536) = @SuppliedVector

SELECT
    t.product_id,
    t.product_name,
    s.distance
FROM
     (
        TABLE = dbo.Products AS t,
        COLUMN = embedding,
        SIMILAR_TO = @query_vector,
        METRIC = 'cosine',
        TOP_N = 10
    ) AS s

Explanation

A vector index is created on the vector-typed embedding column. VECTOR_SEARCH searches that column against the supplied query vector and returns each match’s distance. Ordering by s.distance ranks the nearest vector matches first.

Learn more

Microsoft Fabric SQL database vector search

DP-800: Developing AI-Enabled Database Solutions

AstroTutor

Notes

Readiness

QuestionQ47

Community Discussion