Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    ArrowNotImplementedError
Message:      Cannot write struct type 'model' with no child field to Parquet. Consider adding a dummy child field.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1821, in _prepare_split_single
                  num_examples, num_bytes = writer.finalize()
                                            ^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 781, in finalize
                  self.write_rows_on_file()
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 663, in write_rows_on_file
                  self._write_table(table)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 771, in _write_table
                  self._build_writer(inferred_schema=pa_table.schema)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 812, in _build_writer
                  self.pa_writer = pq.ParquetWriter(
                                   ^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/pyarrow/parquet/core.py", line 1070, in __init__
                  self.writer = _parquet.ParquetWriter(
                                ^^^^^^^^^^^^^^^^^^^^^^^
                File "pyarrow/_parquet.pyx", line 2363, in pyarrow._parquet.ParquetWriter.__cinit__
                File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
                File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
              pyarrow.lib.ArrowNotImplementedError: Cannot write struct type 'model' with no child field to Parquet. Consider adding a dummy child field.
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1832, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

label_map
dict
info
dict
documents
dict
predictions
list
{ "1": "Figure", "2": "Table" }
{ "schema_version": "1.3", "type": "prediction", "created_at": "2026-04-15T08:56:17", "run_id": "unknown-combined-f41f207221", "model": {}, "coordinate_system": { "type": "normalized_xyxy", "range": [ 0, 1 ], "origin": "top_left" } }
{ "doc_id": "3rp_annual_report_2023.pdf", "doc_name": "3rp_annual_report_2023.pdf", "doc_path": "pdf_input/3rp_annual_report_2023.pdf" }
[ { "page_id": "3rp_annual_report_2023.pdf::p003", "doc_id": "3rp_annual_report_2023.pdf", "page_index": 3, "image": { "width_px": 2481, "height_px": 3508, "path": "/data/local-files/?d=unhcr_batch7/3rp_annual_report_2023.pdf_p003.png" }, "objects": [ { "id": "2...
{ "1": "Figure", "2": "Table" }
{ "schema_version": "1.3", "type": "prediction", "created_at": "2026-04-15T08:56:17", "run_id": "unknown-combined-f41f207221", "model": {}, "coordinate_system": { "type": "normalized_xyxy", "range": [ 0, 1 ], "origin": "top_left" } }
{ "doc_id": "1_advocacy_note_mineaction_-_niger_eng.pdf", "doc_name": "1_advocacy_note_mineaction_-_niger_eng.pdf", "doc_path": "pdf_input/1_advocacy_note_mineaction_-_niger_eng.pdf" }
[ { "page_id": "1_advocacy_note_mineaction_-_niger_eng.pdf::p001", "doc_id": "1_advocacy_note_mineaction_-_niger_eng.pdf", "page_index": 1, "image": { "width_px": 2481, "height_px": 3508, "path": "/data/local-files/?d=unhcr_batch9/1_advocacy_note_mineaction_-_niger_eng.pdf_p001.png" ...
{ "1": "Figure", "2": "Table" }
{ "schema_version": "1.3", "type": "prediction", "created_at": "2026-04-15T08:56:17", "run_id": "unknown-combined-f41f207221", "model": {}, "coordinate_system": { "type": "normalized_xyxy", "range": [ 0, 1 ], "origin": "top_left" } }
{ "doc_id": "1_note_plaidoyer_lutte_antimines_-_niger_fr.pdf", "doc_name": "1_note_plaidoyer_lutte_antimines_-_niger_fr.pdf", "doc_path": "pdf_input/1_note_plaidoyer_lutte_antimines_-_niger_fr.pdf" }
[ { "page_id": "1_note_plaidoyer_lutte_antimines_-_niger_fr.pdf::p002", "doc_id": "1_note_plaidoyer_lutte_antimines_-_niger_fr.pdf", "page_index": 2, "image": { "width_px": 2481, "height_px": 3508, "path": "/data/local-files/?d=unhcr_batch4/1_note_plaidoyer_lutte_antimines_-_niger_fr...
{ "1": "Figure", "2": "Table" }
{ "schema_version": "1.3", "type": "prediction", "created_at": "2026-04-15T08:56:17", "run_id": "unknown-combined-f41f207221", "model": {}, "coordinate_system": { "type": "normalized_xyxy", "range": [ 0, 1 ], "origin": "top_left" } }
{ "doc_id": "2_note_danalyse_de_protection_-_retour_de_pdi_a_teguy.pdf", "doc_name": "2_note_danalyse_de_protection_-_retour_de_pdi_a_teguy.pdf", "doc_path": "pdf_input/2_note_danalyse_de_protection_-_retour_de_pdi_a_teguy.pdf" }
[]

Dataset card for data-snapshot

Dataset summary

The data-snapshot dataset is an annotated corpus designed for the evaluation and development of models for extracting data snapshots from PDF documents. A data snapshot is defined as a figure or table that contains quantitative data derived from statistics, indicators, or structured data sources.

Dataset structure

The repository is organized as follows:

ai4data/data-snapshot/
├── annotations/<source>/per_document/*.json    # Contains annotation files per document
├── annotations/<source>/combined/*.json        # Combined annotations into 1 JSON file per source
├── documents/<source>/*.pdf                    # Raw PDFs
├── metadata/<source>/*.json                    # Document-level metadata
├── schemas/data-snapshot-eval-v1.3.schema.json # Provides the schema of the annotation file					
└── README.md

Subsets

  • annotations
    • JSON files that indicate the data snapshots: their object class (Figure / Table) and bounding box locations (in normalized [x1, y1, x2, y2] format, top-left origin)
    • Follows the schema provided in data-snapshot-eval-v1.3.schema.json
    • Provided on a per-document basis or a combined JSON file per source
  • metadata
    • Provided on a per-document basis

Sources

  • UNHCR
  • PRWP (WIP)
  • Refugee (WIP)

Schema

The annotation files follow the Data Snapshot Evaluation Format (v1.3). Below is a simplified, human-readable example of the JSON schema with explanatory comments for each field.

Note: You will notice a top-level field called predictions. In the context of this dataset, this is a misnomer because these are actually human-labeled annotations (ground truth). We use the key predictions because we borrow this schema from the project's evaluation codebase, which uses a unified structure for both ground truth and model predictions.

{
  // Canonical mapping of integer IDs to class names
  "label_map": {
    "1": "Figure",
    "2": "Table"
  },
  
  // High-level metadata about the file
  "info": {
    "schema_version": "1.3",
    "type": "ground_truth",  // Indicates these are human annotations
    "dataset_id": "data-snapshot_unhcr",
    "created_at": "2026-04-17T12:00:00Z",
    "coordinate_system": {
      "type": "normalized_xyxy",
      "range": [0.0, 1.0],  // Bounding boxes are normalized between 0 and 1
      "origin": "top_left"
    }
  },
  
  // List of documents referenced in this file
  "documents": [
    {
      "doc_id": "1_advocacy_note_mineaction_-_niger_eng.pdf",
      "doc_name": "1_advocacy_note_mineaction_-_niger_eng.pdf",
      "doc_path": "pdf_input/1_advocacy_note_mineaction_-_niger_eng.pdf"
    }
  ],
  
  // Per-page container of objects; these contain the ground truth annotations
  "predictions": [
    {
      "page_id": "1_advocacy_note_mineaction_-_niger_eng.pdf::p001",
      "doc_id": "1_advocacy_note_mineaction_-_niger_eng.pdf",
      "page_index": 0,  // 0-indexed page number
      // Image data for Label Studio (ignore this)
      "image": {
        "width_px": 2481,
        "height_px": 3508,
        "path": "images/1_advocacy_note_mineaction_-_niger_eng.pdf_p001.png"
      },
      "objects": [
        {
          "id": "obj_001",
          "label": "Figure",  // Matches a label_map entry
          "bbox": [0.1, 0.2, 0.8, 0.6],  // Normalized [x_min, y_min, x_max, y_max]
        }
      ]
    }
  ]
}

Dataset creation

The annotations were produced through human labeling using Label Studio.

Licensing information

[TBD]

Citation information

[TBD]

Downloads last month
-