|
| 1 | +# Qdrant online store (contrib) |
| 2 | + |
| 3 | +## Description |
| 4 | + |
| 5 | +[Qdrant](http://qdrant.tech) is a vector similarity search engine. It provides a production-ready service with a convenient API to store, search, and manage vectors with additional payload and extended filtering support. It makes it useful for all sorts of neural network or semantic-based matching, faceted search, and other applications. |
| 6 | + |
| 7 | +## Getting started |
| 8 | + |
| 9 | +In order to use this online store, you'll need to run `pip install 'feast[qdrant]'`. |
| 10 | + |
| 11 | +## Example |
| 12 | + |
| 13 | +{% code title="feature_store.yaml" %} |
| 14 | + |
| 15 | +```yaml |
| 16 | +project: my_feature_repo |
| 17 | +registry: data/registry.db |
| 18 | +provider: local |
| 19 | +online_store: |
| 20 | + type: qdrant |
| 21 | + host: localhost |
| 22 | + port: 6333 |
| 23 | + vector_len: 384 |
| 24 | + write_batch_size: 100 |
| 25 | +``` |
| 26 | +
|
| 27 | +{% endcode %} |
| 28 | +
|
| 29 | +The full set of configuration options is available in [QdrantOnlineStoreConfig](https://rtd.feast.dev/en/master/#feast.infra.online_stores.contrib.qdrant.QdrantOnlineStoreConfig). |
| 30 | +
|
| 31 | +## Functionality Matrix |
| 32 | +
|
| 33 | +| | Qdrant | |
| 34 | +| :-------------------------------------------------------- | :------- | |
| 35 | +| write feature values to the online store | yes | |
| 36 | +| read feature values from the online store | yes | |
| 37 | +| update infrastructure (e.g. tables) in the online store | yes | |
| 38 | +| teardown infrastructure (e.g. tables) in the online store | yes | |
| 39 | +| generate a plan of infrastructure changes | no | |
| 40 | +| support for on-demand transforms | yes | |
| 41 | +| readable by Python SDK | yes | |
| 42 | +| readable by Java | no | |
| 43 | +| readable by Go | no | |
| 44 | +| support for entityless feature views | yes | |
| 45 | +| support for concurrent writing to the same key | no | |
| 46 | +| support for ttl (time to live) at retrieval | no | |
| 47 | +| support for deleting expired data | no | |
| 48 | +| collocated by feature view | yes | |
| 49 | +| collocated by feature service | no | |
| 50 | +| collocated by entity key | no | |
| 51 | +
|
| 52 | +To compare this set of functionality against other online stores, please see the full [functionality matrix](overview.md#functionality-matrix). |
| 53 | +
|
| 54 | +## Retrieving online document vectors |
| 55 | +
|
| 56 | +The Qdrant online store supports retrieving document vectors for a given list of entity keys. The document vectors are returned as a dictionary where the key is the entity key and the value is the document vector. The document vector is a dense vector of floats. |
| 57 | +
|
| 58 | +{% code title="python" %} |
| 59 | +
|
| 60 | +```python |
| 61 | +from feast import FeatureStore |
| 62 | + |
| 63 | +feature_store = FeatureStore(repo_path="feature_store.yaml") |
| 64 | + |
| 65 | +query_vector = [1.0, 2.0, 3.0, 4.0, 5.0] |
| 66 | +top_k = 5 |
| 67 | + |
| 68 | +# Retrieve the top k closest features to the query vector |
| 69 | +# Since Qdrant supports multiple vectors per entry, |
| 70 | +# the vector to use can be specified in the repo config. |
| 71 | +# Reference: https://qdrant.tech/documentation/concepts/vectors/#named-vectors |
| 72 | +feature_values = feature_store.retrieve_online_documents( |
| 73 | + feature="my_feature", |
| 74 | + query=query_vector, |
| 75 | + top_k=top_k |
| 76 | +) |
| 77 | +``` |
| 78 | + |
| 79 | +{% endcode %} |
| 80 | + |
| 81 | +These APIs are subject to change in future versions of Feast to improve performance and usability. |
0 commit comments