Reputation: 21
I am learning Qdrant through this repository: https://github.com/qdrant/qdrant-js/blob/master/examples/node-js-basic/index.js
And I have a problem with the return vector. I am using this code.
const res1 = await client.search(collectionName, {
vector: queryVector,
limit: 3,
});
console.log('search result: ', res1);
// prints:
// search result: [
// {
// id: 4,
// version: 3,
// score: 0.99248314,
// payload: { city: [Array] },
// vector: null
// },
Even in the documentation, the vector is null. I added an embedding and inserted several records. The similar search is working great, but I am thinking about how to access the embedding that is stored in the vector.
Upvotes: 2
Views: 750
Reputation: 3258
Qdrant does not return stored vectors for retrieval methods by default.
You can enable this with the with_vector
parameter.
const res1 = await client.search(collectionName, {
vector: queryVector,
limit: 3,
with_vector: true,
});
Documentation: https://qdrant.tech/documentation/concepts/search/#payload-and-vector-in-the-result
Upvotes: 1