Step 2 of 4 · ProjectsTax datasets
product_hscode
11,509 product descriptions mapped to HS codes, with a train and test split; provenance undocumented.
What it does
Dataset of 11,509 short product descriptions each labelled with an HS code, as train and test CSV files with a text column and a numeric label column. The card holds only metadata (MIT, English, text-classification), so nothing on the page says where the pairs come from or which HS edition they follow. The label is a number rather than a code string, so check how codes are represented before use.
Where it fits on the map
Similar projects
Of the same kind, for a shared jurisdiction or tax area. Not a ranking.
- apparel-hs-codesOpen JSON database of apparel and textile HS codes with MFN duty rates for the US, EU, UK and Australia.
- cross_rulings_hts_dataset_for_tariffs18,731 US CBP CROSS rulings paired with HTS codes for tariff classification models (the ATLAS paper).
- DeepCalc tax rate dataOpen tax rate data as JSON and an npm package: US federal income tax, EU VAT, US sales tax for 10 states.
- EU Tax DatasetEU VAT rates for all 27 member states, compiled from European Commission sources.
- EU VAT Rates DataOfficial-source VAT rates for 45 European jurisdictions, machine-readable and checked daily.
- hmrc-exchange-ratesJSON mirror of HMRC's monthly exchange rates for customs and VAT, 2015 to 2026, updated automatically.
Maintain this project?
Link to this page from your README, so readers find the facts, the notes and where it fits. The badge is a small SVG served from this site.
[](https://tax-oss.com/projects/product-hscode-samabos/)