Document reference query input for similarity search. Use existing document's pre-computed features without re-processing. Perfect for "find similar documents" functionality. No inference is performed. Use Cases: - Find similar documents - Reverse image search using indexed images - Document-to-document similarity - Multi-hop similarity chains Examples: Simple document reference: json { \"input_mode\": \"document\", \"document_ref\": { \"collection_id\": \"col_products\", \"document_id\": \"doc_item_123\" } }
| Name | Type | Description | Notes |
|---|---|---|---|
| input_mode | str | Discriminator field. Always 'document' for document reference queries. | [optional] [default to 'document'] |
| document_ref | StageDefsDocumentReference | Reference to existing document's pre-computed features. The system fetches the document's feature vectors for the specified feature_uri and uses them directly without re-processing. Document must exist and have features for the specified feature_uri. |
from mixpeek.models.stage_defs_document_query_input import StageDefsDocumentQueryInput
# TODO update the JSON string below
json = "{}"
# create an instance of StageDefsDocumentQueryInput from a JSON string
stage_defs_document_query_input_instance = StageDefsDocumentQueryInput.from_json(json)
# print the JSON string representation of the object
print(StageDefsDocumentQueryInput.to_json())
# convert the object into a dict
stage_defs_document_query_input_dict = stage_defs_document_query_input_instance.to_dict()
# create an instance of StageDefsDocumentQueryInput from a dict
stage_defs_document_query_input_from_dict = StageDefsDocumentQueryInput.from_dict(stage_defs_document_query_input_dict)