Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
image
imagewidth (px)
850
850

Chart Review Image Pool

Public image pool for chart / no-chart review and pitfall labeling.

Contents

  • images/ — chart images
  • manifest.jsonl — one JSON object per image (provenance + labeling fields)

Current size: 5,806 images (~637 MB), all labeled pending.

Source Count License
Our World in Data grapher PNGs 4,306 CC BY 4.0
ChartComplete 1,500 CC BY 4.0

Manifest fields

Field Meaning
id Stable item id
file_name Relative path under this repo
retrieved_at ISO-8601 UTC retrieval time
source_url Direct download URL used
page_url Human-facing source page when known
wayback_url Wayback Machine snapshot URL if found
license Redistribution license identifier
source_dataset chartcomplete or our_world_in_data
label chart / no_chart (null until labeled)
is_chart Boolean mirror of label (null until labeled)
high_quality High-quality chart flag (null until labeled)
chart_type Type hint when known from source taxonomy
taxonomy_id Active pitfall taxonomy id after labeling
taxonomy_version Taxonomy version after labeling
pitfall_ids Selected pitfall ids (empty until labeled)
labeling_status pending until reviewed

Load locally

from pathlib import Path
import json

rows = [json.loads(line) for line in Path("manifest.jsonl").read_text(encoding="utf-8").splitlines()]
print(rows[0]["id"], rows[0]["file_name"])

After swipe-app export, labeled rows follow:

{"file_name":"images/....png","label":"chart","high_quality":true,"taxonomy_id":null,"taxonomy_version":null,"pitfall_ids":[]}

Attribution

Note

Label fields are intentionally empty so a review app can fill them. ChartComplete provides chart_type folder taxonomy hints; OWID chart_type is null until labeled.

Downloads last month
57