AggregationHit

AggregationHit represents one entity returned as a representative hit inside an AggregationBucket. PyMilvus creates these objects from the server response; applications do not construct them directly.

class pymilvus.AggregationHit

Properties and methods

• pk (int | str | None) -

  The entity primary key.

• score (float) -

  The vector similarity score or distance returned for the entity.

• fields (dict[str, Any]) -

  The requested output fields keyed by field name.

• field_ids() (dict[str, int]) -

  Returns a mapping from each returned field name to its numeric schema field ID.

The fields and field_ids() mappings are copies. Changing them does not mutate the hit object.

Example

bucket = result.agg_buckets[0][0]

for hit in bucket.hits:
    print(hit.pk)
    print(hit.score)
    print(hit.fields)
    print(hit.field_ids())