Datasets:

Ichlibitiche
/

roasterdb-specialty-coffee-sample

license: cc-by-nc-4.0
pretty_name: RoasterDB — Specialty Coffee Dataset (Free Sample)
tags:
  - coffee
  - food
  - e-commerce
  - flavor
  - sca-flavor-wheel
  - recommender-systems
size_categories:
  - n<1K

☕ RoasterDB — Specialty Coffee Dataset (Free Sample)

A free sample of RoasterDB: a structured dataset of specialty-coffee products scraped from the direct storefronts of curated artisan roasters worldwide, with tasting notes normalized to the Specialty Coffee Association (SCA) Flavor Wheel and a source URL on every record so any fact can be re-verified.

This sample contains 100 verified-tier records across 72 roasters — a real taste of the schema and quality.

The full dataset covers 8,000+ products, 280+ roasters, 20+ countries, and 11,000+ SCA flavor mappings (SQLite · CSV · JSON).

Get the Full Dataset

🔗 Official Portal: roasterdb.net — full snapshot $99 one-time
🏆 Kaggle: RoasterDB Specialty Coffee Sample
🔄 Live scraping (self-serve): Specialty Coffee Roaster Scraper on Apify — pay-per-result
☕ Interactive Explorer: Dataset Sample Explorers Space — RoasterDB tab

Key Columns

Column	Description
`product_id`, `title`, `source_roaster`	Product identity
`origin_country`, `origin_region`, `altitude_min/max_meters`	Origin, where stated
`process_method`, `roast_level`, `varietals`	Processing attributes
`weight_grams`, `price_currency`, `price_value`	Unit and sanitized price
`tasting_notes_sca_nodes`	`;` -separated SCA wheel paths, e.g. `Fruity > Berry > Blueberry`
`quality_flag`	`good` (verified tier) or `questionable`
`source_url`, `retrieved_at`, `dataset_version`	Full provenance — every record re-verifiable

Coverage is honest and uneven — storefronts don't all publish farm-level data. Full field coverage numbers are documented at roasterdb.net.

Quick Start

import pandas as pd

df = pd.read_csv("hf://datasets/Ichlibitiche/roasterdb-specialty-coffee-sample/roasterdb_sample.csv")
print(len(df), "coffees from", df["source_roaster"].nunique(), "roasters")
# → 100 coffees from 72 roasters

Use Cases

Coffee subscription & recommendation apps (structured product + flavor data)
Flavor-based search and discovery on the SCA graph
ML / RAG corpora over specialty coffee
Market & assortment research across roasters, origins, and price bands

License

Sample data: CC BY-NC 4.0 — free to use with attribution, non-commercial. The full dataset is available under a commercial license at roasterdb.net. Questions or roaster removal requests: RoasterDB@proton.me.