gcloud beta vector-search collections data-objects batch-search
Stay organized with collections
Save and categorize content based on your preferences.
- NAME
-
- gcloud beta vector-search collections data-objects batch-search - batch search data objects from a Vector Search collection
- SYNOPSIS
-
-
gcloud beta vector-search collections data-objects batch-search--collection=COLLECTION--location=LOCATION--searches-from-file=SEARCHES_FROM_FILE[[--rrf-weights=[WEIGHT,…] :--combine-output-data-fields=[DATA_OUTPUT_FIELD,…]--combine-output-metadata-fields=[METADATA_OUTPUT_FIELD,…]--combine-output-vector-fields=[VECTOR_OUTPUT_FIELD,…]--combine-top-k=COMBINE_TOP_K]] [GCLOUD_WIDE_FLAG …]
-
- DESCRIPTION
-
(BETA)Batch search data objects from a Vector Search collection. Searches can be specified in a JSON file via --searches-from-file. - EXAMPLES
-
To batch search data objects from collection
my-collectionin locationus-central1using searches defined insearches.json, run:gcloudbetavector-searchcollectionsdata-objectsbatch-search--collection=my-collection--location=us-central1--searches-from-file=searches.jsonExample
searches.json:[ { "semanticSearch":{ "searchText":"sci-fi movie", "searchField":"vector1", "topK":10 } }, { "vectorSearch":{ "searchField":"vector2", "topK":5, "vector":{"values":[0.1,0.2,0.3]} } } ]
- REQUIRED FLAGS
-
--collection=COLLECTION- The collection to batch search data objects from.
--location=LOCATION- Location of the collection.
--searches-from-file=SEARCHES_FROM_FILE-
Path to a JSON file containing a list of searches. Each element in list should
be a JSON object representing a Search message, e.g.,
{"semanticSearch": {"searchText": "…", "searchField": "…"}}. Keys must be camelCase as in API definition. Example file content:[ { "vectorSearch":{ "vector":{ "values":[ 1, 2, 3, 4 ] }, "searchField":"genre_embedding", "topK":10, "outputFields":{ "dataFields":[ "director", "genre", "title", "year" ] } } }, { "semanticSearch":{ "searchText":"sci-fi movie", "searchField":"plot_embedding", "taskType":"SEMANTIC_SIMILARITY", "topK":10 } }, { "textSearch":{ "searchText":"movie 1", "dataFieldNames":["title"], "topK":10 } } ]
- OPTIONAL FLAGS
-
- Combine Results Options
-
Ranker
This must be specified.
- RRF Ranker
--rrf-weights=[WEIGHT,…]- RRF weights for combining results.
--combine-output-data-fields=[DATA_OUTPUT_FIELD,…]- List of data fields to include in combined output.
--combine-output-metadata-fields=[METADATA_OUTPUT_FIELD,…]- List of metadata fields to include in combined output.
--combine-output-vector-fields=[VECTOR_OUTPUT_FIELD,…]- List of vector fields to include in combined output.
--combine-top-k=COMBINE_TOP_K- Top K results to return when combining results.
- GCLOUD WIDE FLAGS
-
These flags are available to all commands:
--access-token-file,--account,--billing-project,--configuration,--flags-file,--flatten,--format,--help,--impersonate-service-account,--log-http,--project,--quiet,--trace-token,--user-output-enabled,--verbosity.Run
$ gcloud helpfor details. - NOTES
-
This command is currently in beta and might change without notice. This variant
is also available:
gcloudvector-searchcollectionsdata-objectsbatch-search