StreamingDiskANN index

The StreamingDiskANN index is a graph-based algorithm that uses the DiskANN algorithm. You can read more about it in the blog announcing its release.

To create this index, run:

vec.create_embedding_index(client.TimescaleVectorIndex())

The above command creates the index using smart defaults. There are a number of parameters you could tune to adjust the accuracy/speed trade-off.

The parameters you can set at index build time are:

Parameter name	Description	Default value
num_neighbors	Sets the maximum number of neighbors per node. Higher values increase accuracy but make the graph traversal slower.	50
search_list_size	This is the S parameter used in the greedy search algorithm used during construction. Higher values improve graph quality at the cost of slower index builds.	100
max_alpha	Is the alpha parameter in the algorithm. Higher values improve graph quality at the cost of slower index builds.	1.0

To set these parameters, you could run:

vec.create_embedding_index(client.TimescaleVectorIndex(num_neighbors=50, search_list_size=100, max_alpha=1.0))

You can also set a parameter to control the accuracy vs. query speed trade-off at query time. The parameter is set in the search() function using the query_params argument. You can set the search_list_size(default: 100). This is the number of additional candidates considered during the graph search at query time. Higher values improve query accuracy while making the query slower.

You can specify this value during search as follows:

vec.search([1.0, 9.0], limit=4, query_params=TimescaleVectorIndexParams(search_list_size=10))

To drop the index, run:

vec.drop_embedding_index()