AgentsforBedrockRuntime / Paginator / Rerank
Rerank#
- class AgentsforBedrockRuntime.Paginator.Rerank#
paginator = client.get_paginator('rerank')
- paginate(**kwargs)#
Creates an iterator that will paginate through responses from
AgentsforBedrockRuntime.Client.rerank()
.See also: AWS API Documentation
Request Syntax
response_iterator = paginator.paginate( queries=[ { 'textQuery': { 'text': 'string' }, 'type': 'TEXT' }, ], rerankingConfiguration={ 'bedrockRerankingConfiguration': { 'modelConfiguration': { 'additionalModelRequestFields': { 'string': {...}|[...]|123|123.4|'string'|True|None }, 'modelArn': 'string' }, 'numberOfResults': 123 }, 'type': 'BEDROCK_RERANKING_MODEL' }, sources=[ { 'inlineDocumentSource': { 'jsonDocument': {...}|[...]|123|123.4|'string'|True|None, 'textDocument': { 'text': 'string' }, 'type': 'TEXT'|'JSON' }, 'type': 'INLINE' }, ], PaginationConfig={ 'MaxItems': 123, 'StartingToken': 'string' } )
- Parameters:
queries (list) –
[REQUIRED]
An array of objects, each of which contains information about a query to submit to the reranker model.
(dict) –
Contains information about a query to submit to the reranker model.
textQuery (dict) – [REQUIRED]
Contains information about a text query.
text (string) –
The text of the document.
type (string) – [REQUIRED]
The type of the query.
rerankingConfiguration (dict) –
[REQUIRED]
Contains configurations for reranking.
bedrockRerankingConfiguration (dict) – [REQUIRED]
Contains configurations for an Amazon Bedrock reranker.
modelConfiguration (dict) – [REQUIRED]
Contains configurations for a reranker model.
additionalModelRequestFields (dict) –
A JSON object whose keys are request fields for the model and whose values are values for those fields.
(string) –
(document) –
modelArn (string) – [REQUIRED]
The ARN of the reranker model.
numberOfResults (integer) –
The number of results to return after reranking.
type (string) – [REQUIRED]
The type of reranker that the configurations apply to.
sources (list) –
[REQUIRED]
An array of objects, each of which contains information about the sources to rerank.
(dict) –
Contains information about a source for reranking.
inlineDocumentSource (dict) – [REQUIRED]
Contains an inline definition of a source for reranking.
jsonDocument (document) –
Contains a JSON document to rerank.
textDocument (dict) –
Contains information about a text document to rerank.
text (string) –
The text of the document.
type (string) – [REQUIRED]
The type of document to rerank.
type (string) – [REQUIRED]
The type of the source.
PaginationConfig (dict) –
A dictionary that provides parameters to control pagination.
MaxItems (integer) –
The total number of items to return. If the total number of items available is more than the value specified in max-items then a
NextToken
will be provided in the output that you can use to resume pagination.StartingToken (string) –
A token to specify where to start paginating. This is the
NextToken
from a previous response.
- Return type:
dict
- Returns:
Response Syntax
{ 'results': [ { 'document': { 'jsonDocument': {...}|[...]|123|123.4|'string'|True|None, 'textDocument': { 'text': 'string' }, 'type': 'TEXT'|'JSON' }, 'index': 123, 'relevanceScore': ... }, ], 'NextToken': 'string' }
Response Structure
(dict) –
results (list) –
An array of objects, each of which contains information about the results of reranking.
(dict) –
Contains information about a document that was reranked.
document (dict) –
Contains information about the document.
jsonDocument (document) –
Contains a JSON document to rerank.
textDocument (dict) –
Contains information about a text document to rerank.
text (string) –
The text of the document.
type (string) –
The type of document to rerank.
index (integer) –
The ranking of the document. The lower a number, the higher the document is ranked.
relevanceScore (float) –
The relevance score of the document.
NextToken (string) –
A token to resume pagination.