CtrlK
CommunityDocumentationLog inGet started
Tessl Logo

tessl/pypi-azure-ai-documentintelligence

tessl install tessl/pypi-azure-ai-documentintelligence@1.0.0

Azure AI Document Intelligence client library for Python - a cloud service that uses machine learning to analyze text and structured data from documents

Agent Success

Agent success rate when using this tile

76%

Improvement

Agent success rate improvement when using this tile compared to baseline

1.19x

Baseline

Agent success rate without this tile

64%

rubric.jsonevals/scenario-4/

{
  "context": "Evaluates whether the solution uses azure-ai-documentintelligence batch APIs to launch runs with controlled storage prefixes, resume operations from continuation tokens, and page through batch history. Focuses on correct use of the documented request fields, pollers, and paged iterators rather than custom implementations.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Batch start API",
      "description": "Uses DocumentIntelligenceClient.begin_analyze_batch_documents (or async equivalent) with an AnalyzeBatchDocumentsRequest that supplies model_id plus an AzureBlobContentSource/AzureBlobFileListContentSource and result_container_url; relies on the returned poller for the operation lifecycle.",
      "max_score": 30
    },
    {
      "name": "Prefix/overwrite",
      "description": "Sets result_prefix and overwrite_existing on the AnalyzeBatchDocumentsRequest to control storage paths and conflict behavior, allowing the service to reject overwrites when overwrite_existing is False instead of manually deleting or ignoring existing results.",
      "max_score": 20
    },
    {
      "name": "Continuation resume",
      "description": "Resumes an existing batch using the provided continuation token via get_analyze_batch_result or AnalyzeDocumentLROPoller.from_continuation_token and waits on poller.result() to obtain the final AnalyzeBatchResult rather than starting a new run.",
      "max_score": 20
    },
    {
      "name": "Pagination controls",
      "description": "Lists batch operations with list_analyze_batch_results using skip and top to request specific slices, iterating or by_page as needed to return only that page and capturing the service-provided continuation/next link for further paging.",
      "max_score": 15
    },
    {
      "name": "Operation metadata",
      "description": "Extracts operation_id from poller.details or AnalyzeBatchOperation.operation_id and captures poller.continuation_token() plus succeeded_count/failed_count from AnalyzeBatchResult instead of fabricating identifiers or counts.",
      "max_score": 15
    }
  ]
}

Version

Workspace
tessl
Visibility
Public
Created
Last updated
Describes
pypipkg:pypi/azure-ai-documentintelligence@1.0.x
tile.json