Vector Search Using the Deep Lake Tensor Database
Running Vector Search in the Deep Lake Tensor Database
Last updated
Was this helpful?
Running Vector Search in the Deep Lake Tensor Database
Last updated
Was this helpful?
The REST API is currently in Alpha, and the syntax may change without announcement.
Deep Lake offers a for querying datasets, including vector search. The database is serverless, which simplifies self-hosting and substantially lowers costs. This tutorial demonstrates how to execute embedding search on a Deep Lake dataset using the REST API for the Tensor Database.
A Deep Lake dataset containing embeddings can be created using a variety of APIs, .
Most importantly, when a dataset is stored in the Deep Lake Managed Tensor Database, the dataset dataset_path
and runtime
must be specified accordingly:
In this tutorial, the dataset has already been created .
Let's query our dataset stored in the Managed Tensor Database using the . The steps are:
Define the authentication tokens and search terms
Embed the search search term using OpenAI
Reformat the embedding to an embedding_search
string that can be passed to the REST API request.
Create the query string using . The dataset_path
and embedding_search
are a part of the query string.
Submit the request and print the response data data
Congrats! You performed a vector search using the Deep Lake Managed Database! 🎉