Chart and diagram images paired with the structured data used to render them, spanning seven subject categories from natural sciences to business and finance.
The top entries by volume, with the long tail collapsed into a single figure.
Main category
Every field a delivered record carries, its type, and an example value.
| Field | Type | Description | Example |
|---|---|---|---|
| main_category | string | Top-level subject category from the source taxonomy. | Engineering, Technology & Infrastructure |
| subcategory | string | Subject subcategory within the main category. | IT, Software & Cloud Computing |
| image_count | integer | Images catalogued under this subcategory. | 46500 |
Scroll to see all columns
| Field | Type | Description | Example |
|---|---|---|---|
| image_format | string | Delivered raster or vector image format. | PNG |
| chart_type | string | Visual structure of the rendered image. | bar chart |
| structured_pairs_format | string | Format of the paired structured data the image was rendered from. | JSON |
| rendering_provenance | string | Whether the image was rendered from source data or redrawn from an editorial original. | synthetic-rendered |
Scroll to see all columns
Real rows, exactly as delivered. Copy them and check them.
[ { "main_category": "Engineering, Technology & Infrastructure", "subcategory": "IT, Software & Cloud Computing", "image_count": 46500 }]Sources, acquisition method, and the basis on which the material is held.
Every record passes these stages in order. Each stage states what it removes.
Renders each chart or diagram from its structured source, or ingests a cleared redrawn editorial image.
Source data that fails to render cleanly.
Confirms the structured data and rendered image describe the same values.
Mismatched image/data pairs.
Assigns main category and subcategory via the source taxonomy.
Records the taxonomy cannot place.
Compares perceptual hashes across the catalog.
Near-identical renders.
Flags low-resolution or overlapping-label images.
Images that fail the legibility threshold.
Named benchmarks checked for overlap, and the method used to check them.
Formats, access mechanisms, and how revisions are published.
What this dataset does not cover. Read this before you evaluate it.
We share sample records on request. Tell us your training stage and intended use and we will send an evaluation copy where the use fits.
Request sample records