Update indexing.md

This commit is contained in:
davidmyriel
2024-05-19 19:42:48 -07:00
parent 1b81245fab
commit 4f4bc6bf6a
@@ -465,16 +465,14 @@ performance.
*Available as of v1.7.0*
Sparse vectors in Qdrant are indexed with a special data structure, optimized for vectors with a high proportion of zeroes.
In some ways, it is similar to the inverted index, used in text search engines.
Sparse vectors in Qdrant are indexed with a special data structure, which is optimized for vectors that have a high proportion of zeroes. In some ways, this indexing method is similar to the inverted index, which is used in text search engines.
The sparse vector index in Qdrant is exact, meaning it does not use any approximation algorithms.
- A sparse vector index in Qdrant is exact, meaning it does not use any approximation algorithms.
- All sparse vectors added to the collection are immediately indexed in the mutable version of a sparse index.
All sparse vectors added to the collection are immediately indexed in the mutable version of a sparse index.
Qdrant, however, allows you to also benefit from a more compact and efficient immutable sparse index, which is constructed during the same optimization process as the dense vector index.
That is especially useful for collections, which have both dense and sparse vectors stored.
With Qdrant, you can benefit from a more compact and efficient immutable sparse index, which is constructed during the same optimization process as the dense vector index.
This approach is particularly useful for collections storing both dense and sparse vectors.
To configure a sparse vector index, create a collection with the following parameters:
@@ -595,17 +593,14 @@ await client.CreateCollectionAsync(
);
```
Some important parameters of the sparse index are:
The following parameters may affect performance:
- `on_disk: true` - the index is stored on disk, which lets you save memory, but may also slow down search performance.
- If `on_disk` is set to `false`, the sparse index is still persisted on disk, but it is also loaded into memory for faster search.
- `on_disk: true` - The index is stored on disk, which lets you save memory. This may slow down search performance.
- `on_disk: false` - The index is still persisted on disk, but it is also loaded into memory for faster search.
<!-- Modifier explanation -->
Unlike a dense vector index, a sparse vector index does not require a pre-defined vector size. It automatically adjusts to the size of the vectors added to the collection.
Unlike dense vector index, a sparse vector index does not require pre-defined size of the vector. It is automatically adjusted to the size of the vectors added to the collection.
**Note:** The sparse vector index only supports dot-product similarity searches. It does not support other distance metrics.
**Note:** A sparse vector index only supports dot-product similarity searches. It does not support other distance metrics.
## Filtrable Index