Vector Search

These dataclasses are used in the SDK to represent API requests and responses for services in the databricks.sdk.service.vectorsearch module.

class databricks.sdk.service.vectorsearch.ColumnInfo(name: 'Optional[str]' = None, type_text: 'Optional[str]' = None)

name: str | None = None

Name of the column.

type_text: str | None = None

Data type of the column (e.g., “string”, “int”, “array<float>”)

as_dict() → dict

Serializes the ColumnInfo into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ColumnInfo into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ColumnInfo

Deserializes the ColumnInfo from a dictionary.

class databricks.sdk.service.vectorsearch.CustomTag(key: 'str', value: 'Optional[str]' = None)

key: str

Key field for an AI Search endpoint tag.

value: str | None = None

[Optional] Value field for an AI Search endpoint tag.

as_dict() → dict

Serializes the CustomTag into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the CustomTag into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → CustomTag

Deserializes the CustomTag from a dictionary.

class databricks.sdk.service.vectorsearch.DeleteDataResult(failed_primary_keys: 'Optional[List[str]]' = None, success_row_count: 'Optional[int]' = None)

failed_primary_keys: List[str] | None = None

List of primary keys for rows that failed to process.

success_row_count: int | None = None

Count of successfully processed rows.

as_dict() → dict

Serializes the DeleteDataResult into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeleteDataResult into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeleteDataResult

Deserializes the DeleteDataResult from a dictionary.

class databricks.sdk.service.vectorsearch.DeleteDataStatus

FAILURE = "FAILURE"

PARTIAL_SUCCESS = "PARTIAL_SUCCESS"

SUCCESS = "SUCCESS"

class databricks.sdk.service.vectorsearch.DeleteDataVectorIndexResponse(result: 'Optional[DeleteDataResult]' = None, status: 'Optional[DeleteDataStatus]' = None)

result: DeleteDataResult | None = None

Result of the upsert or delete operation.

status: DeleteDataStatus | None = None

Status of the delete operation.

as_dict() → dict

Serializes the DeleteDataVectorIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeleteDataVectorIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeleteDataVectorIndexResponse

Deserializes the DeleteDataVectorIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.DeleteEndpointResponse

as_dict() → dict

Serializes the DeleteEndpointResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeleteEndpointResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeleteEndpointResponse

Deserializes the DeleteEndpointResponse from a dictionary.

class databricks.sdk.service.vectorsearch.DeleteIndexResponse

as_dict() → dict

Serializes the DeleteIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeleteIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeleteIndexResponse

Deserializes the DeleteIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.DeltaSyncVectorIndexSpecRequest(columns_to_index: 'Optional[List[str]]' = None, columns_to_sync: 'Optional[List[str]]' = None, embedding_source_columns: 'Optional[List[EmbeddingSourceColumn]]' = None, embedding_vector_columns: 'Optional[List[EmbeddingVectorColumn]]' = None, embedding_writeback_table: 'Optional[str]' = None, pipeline_type: 'Optional[PipelineType]' = None, source_table: 'Optional[str]' = None)

columns_to_index: List[str] | None = None

[Optional] Alias for columns_to_sync. Select the columns to include in the vector index. If you leave this field blank, all columns from the source table are included. The primary key column and embedding source column or embedding vector column are always included. Only one of columns_to_sync or columns_to_index may be specified.

columns_to_sync: List[str] | None = None

[Optional] Select the columns to sync with the vector index. If you leave this field blank, all columns from the source table are synced with the index. The primary key column and embedding source column or embedding vector column are always synced.

embedding_source_columns: List[EmbeddingSourceColumn] | None = None

The columns that contain the embedding source.

embedding_vector_columns: List[EmbeddingVectorColumn] | None = None

The columns that contain the embedding vectors.

embedding_writeback_table: str | None = None

[Optional] Name of the Delta table to sync the vector index contents and computed embeddings to.

pipeline_type: PipelineType | None = None

Pipeline execution mode. - TRIGGERED: If the pipeline uses the triggered execution mode, the system stops processing after successfully refreshing the source table in the pipeline once, ensuring the table is updated based on the data available when the update started. - CONTINUOUS: If the pipeline uses continuous execution, the pipeline processes new data as it arrives in the source table to keep vector index fresh.

source_table: str | None = None

The name of the source table.

as_dict() → dict

Serializes the DeltaSyncVectorIndexSpecRequest into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeltaSyncVectorIndexSpecRequest into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeltaSyncVectorIndexSpecRequest

Deserializes the DeltaSyncVectorIndexSpecRequest from a dictionary.

class databricks.sdk.service.vectorsearch.DeltaSyncVectorIndexSpecResponse(columns_to_index: 'Optional[List[str]]' = None, columns_to_sync: 'Optional[List[str]]' = None, embedding_source_columns: 'Optional[List[EmbeddingSourceColumn]]' = None, embedding_vector_columns: 'Optional[List[EmbeddingVectorColumn]]' = None, embedding_writeback_table: 'Optional[str]' = None, pipeline_id: 'Optional[str]' = None, pipeline_type: 'Optional[PipelineType]' = None, source_table: 'Optional[str]' = None)

columns_to_index: List[str] | None = None

[Optional] Alias for columns_to_sync. Select the columns to include in the vector index. If you leave this field blank, all columns from the source table are included. The primary key column and embedding source column or embedding vector column are always included. Only one of columns_to_sync or columns_to_index may be specified.

columns_to_sync: List[str] | None = None

[Optional] Select the columns to sync with the vector index. If you leave this field blank, all columns from the source table are synced with the index. The primary key column and embedding source column or embedding vector column are always synced.

embedding_source_columns: List[EmbeddingSourceColumn] | None = None

The columns that contain the embedding source.

embedding_vector_columns: List[EmbeddingVectorColumn] | None = None

The columns that contain the embedding vectors.

embedding_writeback_table: str | None = None

[Optional] Name of the Delta table to sync the vector index contents and computed embeddings to.

pipeline_id: str | None = None

The ID of the pipeline that is used to sync the index.

pipeline_type: PipelineType | None = None

Pipeline execution mode. - TRIGGERED: If the pipeline uses the triggered execution mode, the system stops processing after successfully refreshing the source table in the pipeline once, ensuring the table is updated based on the data available when the update started. - CONTINUOUS: If the pipeline uses continuous execution, the pipeline processes new data as it arrives in the source table to keep vector index fresh.

source_table: str | None = None

The name of the source table.

as_dict() → dict

Serializes the DeltaSyncVectorIndexSpecResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DeltaSyncVectorIndexSpecResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DeltaSyncVectorIndexSpecResponse

Deserializes the DeltaSyncVectorIndexSpecResponse from a dictionary.

class databricks.sdk.service.vectorsearch.DirectAccessVectorIndexSpec(embedding_source_columns: 'Optional[List[EmbeddingSourceColumn]]' = None, embedding_vector_columns: 'Optional[List[EmbeddingVectorColumn]]' = None, schema_json: 'Optional[str]' = None)

embedding_source_columns: List[EmbeddingSourceColumn] | None = None

The columns that contain the embedding source. The format should be array[double].

embedding_vector_columns: List[EmbeddingVectorColumn] | None = None

The columns that contain the embedding vectors. The format should be array[double].

schema_json: str | None = None

The schema of the index in JSON format. Supported types are integer, long, float, double, boolean, string, date, timestamp. Supported types for vector column: array<float>, array<double>,`.

as_dict() → dict

Serializes the DirectAccessVectorIndexSpec into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the DirectAccessVectorIndexSpec into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → DirectAccessVectorIndexSpec

Deserializes the DirectAccessVectorIndexSpec from a dictionary.

class databricks.sdk.service.vectorsearch.EmbeddingSourceColumn(embedding_model_endpoint_name: 'Optional[str]' = None, model_endpoint_name_for_query: 'Optional[str]' = None, name: 'Optional[str]' = None)

embedding_model_endpoint_name: str | None = None

Name of the embedding model endpoint, used by default for both ingestion and querying.

model_endpoint_name_for_query: str | None = None

Name of the embedding model endpoint which, if specified, is used for querying (not ingestion).

name: str | None = None

Name of the column

as_dict() → dict

Serializes the EmbeddingSourceColumn into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the EmbeddingSourceColumn into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → EmbeddingSourceColumn

Deserializes the EmbeddingSourceColumn from a dictionary.

class databricks.sdk.service.vectorsearch.EmbeddingVectorColumn(embedding_dimension: 'Optional[int]' = None, name: 'Optional[str]' = None)

embedding_dimension: int | None = None

Dimension of the embedding vector

name: str | None = None

Name of the column

as_dict() → dict

Serializes the EmbeddingVectorColumn into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the EmbeddingVectorColumn into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → EmbeddingVectorColumn

Deserializes the EmbeddingVectorColumn from a dictionary.

class databricks.sdk.service.vectorsearch.EndpointInfo(budget_policy_id: 'Optional[str]' = None, creation_timestamp: 'Optional[int]' = None, creator: 'Optional[str]' = None, custom_tags: 'Optional[List[CustomTag]]' = None, effective_budget_policy_id: 'Optional[str]' = None, endpoint_status: 'Optional[EndpointStatus]' = None, endpoint_type: 'Optional[EndpointType]' = None, id: 'Optional[str]' = None, last_updated_timestamp: 'Optional[int]' = None, last_updated_user: 'Optional[str]' = None, name: 'Optional[str]' = None, num_indexes: 'Optional[int]' = None, scaling_info: 'Optional[EndpointScalingInfo]' = None)

budget_policy_id: str | None = None

The user-selected budget policy id for the endpoint.

creation_timestamp: int | None = None

Timestamp of endpoint creation

creator: str | None = None

Creator of the endpoint

custom_tags: List[CustomTag] | None = None

The custom tags assigned to the endpoint

effective_budget_policy_id: str | None = None

The budget policy id applied to the endpoint

endpoint_status: EndpointStatus | None = None

Current status of the endpoint

endpoint_type: EndpointType | None = None

Type of endpoint

id: str | None = None

Unique identifier of the endpoint

last_updated_timestamp: int | None = None

Timestamp of last update to the endpoint

last_updated_user: str | None = None

User who last updated the endpoint

name: str | None = None

Name of the AI Search endpoint

num_indexes: int | None = None

Number of indexes on the endpoint

scaling_info: EndpointScalingInfo | None = None

Scaling information for the endpoint

as_dict() → dict

Serializes the EndpointInfo into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the EndpointInfo into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → EndpointInfo

Deserializes the EndpointInfo from a dictionary.

class databricks.sdk.service.vectorsearch.EndpointScalingInfo(requested_target_qps: 'Optional[int]' = None, state: 'Optional[ScalingChangeState]' = None)

requested_target_qps: int | None = None

The requested QPS target for the endpoint. Best-effort; the system does not guarantee this QPS will be achieved.

state: ScalingChangeState | None = None

The current state of the scaling change request.

as_dict() → dict

Serializes the EndpointScalingInfo into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the EndpointScalingInfo into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → EndpointScalingInfo

Deserializes the EndpointScalingInfo from a dictionary.

class databricks.sdk.service.vectorsearch.EndpointStatus(message: str | None = None, state: EndpointStatusState | None = None)

Status information of an endpoint

message: str | None = None

Additional status message

state: EndpointStatusState | None = None

Current state of the endpoint

as_dict() → dict

Serializes the EndpointStatus into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the EndpointStatus into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → EndpointStatus

Deserializes the EndpointStatus from a dictionary.

class databricks.sdk.service.vectorsearch.EndpointStatusState

Current state of the endpoint

DELETED = "DELETED"

OFFLINE = "OFFLINE"

ONLINE = "ONLINE"

PROVISIONING = "PROVISIONING"

RED_STATE = "RED_STATE"

YELLOW_STATE = "YELLOW_STATE"

class databricks.sdk.service.vectorsearch.EndpointType

Type of endpoint.

STANDARD = "STANDARD"

STORAGE_OPTIMIZED = "STORAGE_OPTIMIZED"

class databricks.sdk.service.vectorsearch.GetVectorSearchEndpointPermissionLevelsResponse(permission_levels: 'Optional[List[VectorSearchEndpointPermissionsDescription]]' = None)

permission_levels: List[VectorSearchEndpointPermissionsDescription] | None = None

Specific permission levels

as_dict() → dict

Serializes the GetVectorSearchEndpointPermissionLevelsResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the GetVectorSearchEndpointPermissionLevelsResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → GetVectorSearchEndpointPermissionLevelsResponse

Deserializes the GetVectorSearchEndpointPermissionLevelsResponse from a dictionary.

class databricks.sdk.service.vectorsearch.IndexSubtype

The subtype of the AI Search index, determining the indexing and retrieval strategy. - VECTOR: Not supported. Use HYBRID instead. - FULL_TEXT: An index that uses full-text search without vector embeddings. - HYBRID: An index that uses vector embeddings for similarity search and hybrid search.

FULL_TEXT = "FULL_TEXT"

HYBRID = "HYBRID"

VECTOR = "VECTOR"

class databricks.sdk.service.vectorsearch.ListEndpointResponse(endpoints: 'Optional[List[EndpointInfo]]' = None, next_page_token: 'Optional[str]' = None)

endpoints: List[EndpointInfo] | None = None

An array of Endpoint objects

next_page_token: str | None = None

A token that can be used to get the next page of results. If not present, there are no more results to show.

as_dict() → dict

Serializes the ListEndpointResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ListEndpointResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ListEndpointResponse

Deserializes the ListEndpointResponse from a dictionary.

class databricks.sdk.service.vectorsearch.ListValue(values: 'Optional[List[Value]]' = None)

values: List[Value] | None = None

Repeated field of dynamically typed values.

as_dict() → dict

Serializes the ListValue into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ListValue into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ListValue

Deserializes the ListValue from a dictionary.

class databricks.sdk.service.vectorsearch.ListVectorIndexesResponse(next_page_token: 'Optional[str]' = None, vector_indexes: 'Optional[List[MiniVectorIndex]]' = None)

next_page_token: str | None = None

A token that can be used to get the next page of results. If not present, there are no more results to show.

vector_indexes: List[MiniVectorIndex] | None = None

as_dict() → dict

Serializes the ListVectorIndexesResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ListVectorIndexesResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ListVectorIndexesResponse

Deserializes the ListVectorIndexesResponse from a dictionary.

class databricks.sdk.service.vectorsearch.MapStringValueEntry(key: str | None = None, value: Value | None = None)

Key-value pair.

key: str | None = None

Column name.

value: Value | None = None

Column value, nullable.

as_dict() → dict

Serializes the MapStringValueEntry into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the MapStringValueEntry into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → MapStringValueEntry

Deserializes the MapStringValueEntry from a dictionary.

class databricks.sdk.service.vectorsearch.Metric(labels: List[MetricLabel] | None = None, name: str | None = None, percentile: float | None = None)

Metric specification

labels: List[MetricLabel] | None = None

Metric labels

name: str | None = None

Metric name

percentile: float | None = None

Percentile for the metric

as_dict() → dict

Serializes the Metric into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the Metric into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → Metric

Deserializes the Metric from a dictionary.

class databricks.sdk.service.vectorsearch.MetricLabel(name: str | None = None, value: str | None = None)

Label for a metric

name: str | None = None

Label name

value: str | None = None

Label value

as_dict() → dict

Serializes the MetricLabel into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the MetricLabel into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → MetricLabel

Deserializes the MetricLabel from a dictionary.

class databricks.sdk.service.vectorsearch.MetricValue(timestamp: int | None = None, value: float | None = None)

Single metric value at a specific timestamp

timestamp: int | None = None

Timestamp of the metric value (milliseconds since epoch)

value: float | None = None

Metric value

as_dict() → dict

Serializes the MetricValue into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the MetricValue into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → MetricValue

Deserializes the MetricValue from a dictionary.

class databricks.sdk.service.vectorsearch.MetricValues(metric: Metric | None = None, values: List[MetricValue] | None = None)

Collection of metric values for a specific metric

metric: Metric | None = None

Metric specification

values: List[MetricValue] | None = None

Time series of metric values

as_dict() → dict

Serializes the MetricValues into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the MetricValues into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → MetricValues

Deserializes the MetricValues from a dictionary.

class databricks.sdk.service.vectorsearch.MiniVectorIndex(creator: 'Optional[str]' = None, endpoint_name: 'Optional[str]' = None, index_subtype: 'Optional[IndexSubtype]' = None, index_type: 'Optional[VectorIndexType]' = None, name: 'Optional[str]' = None, primary_key: 'Optional[str]' = None)

creator: str | None = None

The user who created the index.

endpoint_name: str | None = None

Name of the endpoint associated with the index

index_subtype: IndexSubtype | None = None

The subtype of the index.

index_type: VectorIndexType | None = None

name: str | None = None

Name of the index

primary_key: str | None = None

Primary key of the index

as_dict() → dict

Serializes the MiniVectorIndex into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the MiniVectorIndex into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → MiniVectorIndex

Deserializes the MiniVectorIndex from a dictionary.

class databricks.sdk.service.vectorsearch.PatchEndpointBudgetPolicyResponse(budget_policy_id: 'Optional[str]' = None, effective_budget_policy_id: 'Optional[str]' = None)

budget_policy_id: str | None = None

effective_budget_policy_id: str | None = None

The budget policy applied to the AI Search endpoint.

as_dict() → dict

Serializes the PatchEndpointBudgetPolicyResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the PatchEndpointBudgetPolicyResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → PatchEndpointBudgetPolicyResponse

Deserializes the PatchEndpointBudgetPolicyResponse from a dictionary.

class databricks.sdk.service.vectorsearch.PipelineType

Pipeline execution mode. - TRIGGERED: If the pipeline uses the triggered execution mode, the system stops processing after successfully refreshing the source table in the pipeline once, ensuring the table is updated based on the data available when the update started. - CONTINUOUS: If the pipeline uses continuous execution, the pipeline processes new data as it arrives in the source table to keep vector index fresh.

CONTINUOUS = "CONTINUOUS"

TRIGGERED = "TRIGGERED"

class databricks.sdk.service.vectorsearch.QueryVectorIndexResponse(manifest: 'Optional[ResultManifest]' = None, next_page_token: 'Optional[str]' = None, result: 'Optional[ResultData]' = None)

manifest: ResultManifest | None = None

Metadata about the result set.

next_page_token: str | None = None

[Optional] Token that can be used in QueryVectorIndexNextPage API to get next page of results. If more than 1000 results satisfy the query, they are returned in groups of 1000. Empty value means no more results. The maximum number of results that can be returned is 10,000.

result: ResultData | None = None

Data returned in the query result.

as_dict() → dict

Serializes the QueryVectorIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the QueryVectorIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → QueryVectorIndexResponse

Deserializes the QueryVectorIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.RerankerConfig(model: 'Optional[str]' = None, parameters: 'Optional[RerankerConfigRerankerParameters]' = None)

model: str | None = None

Reranker identifier: - When model_type=BASE/UNSPECIFIED: must be “databricks_reranker”. - When model_type=FINETUNED: the Model Serving endpoint name hosting a finetuned reranker.

parameters: RerankerConfigRerankerParameters | None = None

Parameters that control how the reranker processes the query results.

as_dict() → dict

Serializes the RerankerConfig into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the RerankerConfig into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → RerankerConfig

Deserializes the RerankerConfig from a dictionary.

class databricks.sdk.service.vectorsearch.RerankerConfigRerankerParameters(columns_to_rerank: 'Optional[List[str]]' = None)

columns_to_rerank: List[str] | None = None

as_dict() → dict

Serializes the RerankerConfigRerankerParameters into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the RerankerConfigRerankerParameters into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → RerankerConfigRerankerParameters

Deserializes the RerankerConfigRerankerParameters from a dictionary.

class databricks.sdk.service.vectorsearch.ResultData(data_array: List[List[str]] | None = None, row_count: int | None = None)

Data returned in the query result.

data_array: List[List[str]] | None = None

Data rows returned in the query.

row_count: int | None = None

Number of rows in the result set.

as_dict() → dict

Serializes the ResultData into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ResultData into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ResultData

Deserializes the ResultData from a dictionary.

class databricks.sdk.service.vectorsearch.ResultManifest(column_count: int | None = None, columns: List[ColumnInfo] | None = None)

Metadata about the result set.

column_count: int | None = None

Number of columns in the result set.

columns: List[ColumnInfo] | None = None

Information about each column in the result set.

as_dict() → dict

Serializes the ResultManifest into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ResultManifest into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ResultManifest

Deserializes the ResultManifest from a dictionary.

class databricks.sdk.service.vectorsearch.RetrieveUserVisibleMetricsResponse(metric_values: List[MetricValues] | None = None, next_page_token: str | None = None)

Response containing user-visible metrics

metric_values: List[MetricValues] | None = None

Collection of metric values

next_page_token: str | None = None

A token that can be used to get the next page of results. If not present, there are no more results to show.

as_dict() → dict

Serializes the RetrieveUserVisibleMetricsResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the RetrieveUserVisibleMetricsResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → RetrieveUserVisibleMetricsResponse

Deserializes the RetrieveUserVisibleMetricsResponse from a dictionary.

class databricks.sdk.service.vectorsearch.ScalingChangeState

SCALING_CHANGE_APPLIED = "SCALING_CHANGE_APPLIED"

SCALING_CHANGE_IN_PROGRESS = "SCALING_CHANGE_IN_PROGRESS"

SCALING_CHANGE_UNSPECIFIED = "SCALING_CHANGE_UNSPECIFIED"

class databricks.sdk.service.vectorsearch.ScanVectorIndexResponse(data: List[Struct] | None = None, last_primary_key: str | None = None)

Response to a scan vector index request.

data: List[Struct] | None = None

List of data entries

last_primary_key: str | None = None

Primary key of the last entry.

as_dict() → dict

Serializes the ScanVectorIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the ScanVectorIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → ScanVectorIndexResponse

Deserializes the ScanVectorIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.Struct(fields: 'Optional[List[MapStringValueEntry]]' = None)

fields: List[MapStringValueEntry] | None = None

Data entry, corresponding to a row in a vector index.

as_dict() → dict

Serializes the Struct into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the Struct into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → Struct

Deserializes the Struct from a dictionary.

class databricks.sdk.service.vectorsearch.SyncIndexResponse

as_dict() → dict

Serializes the SyncIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the SyncIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → SyncIndexResponse

Deserializes the SyncIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.UpdateEndpointCustomTagsResponse(custom_tags: 'Optional[List[CustomTag]]' = None, name: 'Optional[str]' = None)

custom_tags: List[CustomTag] | None = None

All the custom tags that are applied to the AI Search endpoint.

name: str | None = None

The name of the AI Search endpoint whose custom tags were updated.

as_dict() → dict

Serializes the UpdateEndpointCustomTagsResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the UpdateEndpointCustomTagsResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → UpdateEndpointCustomTagsResponse

Deserializes the UpdateEndpointCustomTagsResponse from a dictionary.

class databricks.sdk.service.vectorsearch.UpsertDataResult(failed_primary_keys: 'Optional[List[str]]' = None, success_row_count: 'Optional[int]' = None)

failed_primary_keys: List[str] | None = None

List of primary keys for rows that failed to process.

success_row_count: int | None = None

Count of successfully processed rows.

as_dict() → dict

Serializes the UpsertDataResult into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the UpsertDataResult into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → UpsertDataResult

Deserializes the UpsertDataResult from a dictionary.

class databricks.sdk.service.vectorsearch.UpsertDataStatus

FAILURE = "FAILURE"

PARTIAL_SUCCESS = "PARTIAL_SUCCESS"

SUCCESS = "SUCCESS"

class databricks.sdk.service.vectorsearch.UpsertDataVectorIndexResponse(result: 'Optional[UpsertDataResult]' = None, status: 'Optional[UpsertDataStatus]' = None)

result: UpsertDataResult | None = None

Result of the upsert or delete operation.

status: UpsertDataStatus | None = None

Status of the upsert operation.

as_dict() → dict

Serializes the UpsertDataVectorIndexResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the UpsertDataVectorIndexResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → UpsertDataVectorIndexResponse

Deserializes the UpsertDataVectorIndexResponse from a dictionary.

class databricks.sdk.service.vectorsearch.Value(bool_value: 'Optional[bool]' = None, list_value: 'Optional[ListValue]' = None, number_value: 'Optional[float]' = None, string_value: 'Optional[str]' = None, struct_value: 'Optional[Struct]' = None)

bool_value: bool | None = None

list_value: ListValue | None = None

number_value: float | None = None

string_value: str | None = None

struct_value: Struct | None = None

as_dict() → dict

Serializes the Value into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the Value into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → Value

Deserializes the Value from a dictionary.

class databricks.sdk.service.vectorsearch.VectorIndex(creator: 'Optional[str]' = None, delta_sync_index_spec: 'Optional[DeltaSyncVectorIndexSpecResponse]' = None, direct_access_index_spec: 'Optional[DirectAccessVectorIndexSpec]' = None, endpoint_name: 'Optional[str]' = None, index_subtype: 'Optional[IndexSubtype]' = None, index_type: 'Optional[VectorIndexType]' = None, name: 'Optional[str]' = None, primary_key: 'Optional[str]' = None, status: 'Optional[VectorIndexStatus]' = None)

creator: str | None = None

The user who created the index.

delta_sync_index_spec: DeltaSyncVectorIndexSpecResponse | None = None

direct_access_index_spec: DirectAccessVectorIndexSpec | None = None

endpoint_name: str | None = None

Name of the endpoint associated with the index

index_subtype: IndexSubtype | None = None

The subtype of the index.

index_type: VectorIndexType | None = None

name: str | None = None

Name of the index

primary_key: str | None = None

Primary key of the index

status: VectorIndexStatus | None = None

as_dict() → dict

Serializes the VectorIndex into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorIndex into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorIndex

Deserializes the VectorIndex from a dictionary.

class databricks.sdk.service.vectorsearch.VectorIndexStatus(index_url: 'Optional[str]' = None, indexed_row_count: 'Optional[int]' = None, message: 'Optional[str]' = None, ready: 'Optional[bool]' = None)

index_url: str | None = None

Index API Url to be used to perform operations on the index

indexed_row_count: int | None = None

Number of rows indexed

message: str | None = None

Message associated with the index status

ready: bool | None = None

Whether the index is ready for search

as_dict() → dict

Serializes the VectorIndexStatus into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorIndexStatus into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorIndexStatus

Deserializes the VectorIndexStatus from a dictionary.

class databricks.sdk.service.vectorsearch.VectorIndexType

There are 2 types of AI Search indexes: - DELTA_SYNC: An index that automatically syncs with a source Delta Table, automatically and incrementally updating the index as the underlying data in the Delta Table changes. - DIRECT_ACCESS: An index that supports direct read and write of vectors and metadata through our REST and SDK APIs. With this model, the user manages index updates.

DELTA_SYNC = "DELTA_SYNC"

DIRECT_ACCESS = "DIRECT_ACCESS"

class databricks.sdk.service.vectorsearch.VectorSearchEndpointAccessControlRequest(group_name: 'Optional[str]' = None, permission_level: 'Optional[VectorSearchEndpointPermissionLevel]' = None, service_principal_name: 'Optional[str]' = None, user_name: 'Optional[str]' = None)

group_name: str | None = None

name of the group

permission_level: VectorSearchEndpointPermissionLevel | None = None

service_principal_name: str | None = None

application ID of a service principal

user_name: str | None = None

name of the user

as_dict() → dict

Serializes the VectorSearchEndpointAccessControlRequest into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorSearchEndpointAccessControlRequest into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorSearchEndpointAccessControlRequest

Deserializes the VectorSearchEndpointAccessControlRequest from a dictionary.

class databricks.sdk.service.vectorsearch.VectorSearchEndpointAccessControlResponse(all_permissions: 'Optional[List[VectorSearchEndpointPermission]]' = None, display_name: 'Optional[str]' = None, group_name: 'Optional[str]' = None, service_principal_name: 'Optional[str]' = None, user_name: 'Optional[str]' = None)

all_permissions: List[VectorSearchEndpointPermission] | None = None

All permissions.

display_name: str | None = None

Display name of the user or service principal.

group_name: str | None = None

name of the group

service_principal_name: str | None = None

Name of the service principal.

user_name: str | None = None

name of the user

as_dict() → dict

Serializes the VectorSearchEndpointAccessControlResponse into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorSearchEndpointAccessControlResponse into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorSearchEndpointAccessControlResponse

Deserializes the VectorSearchEndpointAccessControlResponse from a dictionary.

class databricks.sdk.service.vectorsearch.VectorSearchEndpointPermission(inherited: 'Optional[bool]' = None, inherited_from_object: 'Optional[List[str]]' = None, permission_level: 'Optional[VectorSearchEndpointPermissionLevel]' = None)

inherited: bool | None = None

inherited_from_object: List[str] | None = None

permission_level: VectorSearchEndpointPermissionLevel | None = None

as_dict() → dict

Serializes the VectorSearchEndpointPermission into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorSearchEndpointPermission into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorSearchEndpointPermission

Deserializes the VectorSearchEndpointPermission from a dictionary.

class databricks.sdk.service.vectorsearch.VectorSearchEndpointPermissionLevel

Permission level

CAN_CREATE = "CAN_CREATE"

CAN_MANAGE = "CAN_MANAGE"

CAN_USE = "CAN_USE"

class databricks.sdk.service.vectorsearch.VectorSearchEndpointPermissions(access_control_list: 'Optional[List[VectorSearchEndpointAccessControlResponse]]' = None, object_id: 'Optional[str]' = None, object_type: 'Optional[str]' = None)

access_control_list: List[VectorSearchEndpointAccessControlResponse] | None = None

object_id: str | None = None

object_type: str | None = None

as_dict() → dict

Serializes the VectorSearchEndpointPermissions into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorSearchEndpointPermissions into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorSearchEndpointPermissions

Deserializes the VectorSearchEndpointPermissions from a dictionary.

class databricks.sdk.service.vectorsearch.VectorSearchEndpointPermissionsDescription(description: 'Optional[str]' = None, permission_level: 'Optional[VectorSearchEndpointPermissionLevel]' = None)

description: str | None = None

permission_level: VectorSearchEndpointPermissionLevel | None = None

as_dict() → dict

Serializes the VectorSearchEndpointPermissionsDescription into a dictionary suitable for use as a JSON request body.

as_shallow_dict() → dict

Serializes the VectorSearchEndpointPermissionsDescription into a shallow dictionary of its immediate attributes.

classmethod from_dict(d: Dict[str, Any]) → VectorSearchEndpointPermissionsDescription

Deserializes the VectorSearchEndpointPermissionsDescription from a dictionary.