eventthing/venue-library

EVENTTHING Venue Library v1.0.1

An open, machine-readable library of the venues where brand experiences happen across Asia. Data CC BY 4.0 · code MIT · published by EVENTTHING.

• Source repository: https://github.com/eventthing-ai/venue-library
• Release: https://github.com/eventthing-ai/venue-library/releases/tag/v1.0.1
• Full documentation (field dictionary, schema, validator): in this dataset under docs/.

Eight views

View	Rows	File
Hotel	3,038	data/hotels.public.csv
All Venue	3,549	data/venues.public.csv
Shop Mall	2,557	subset of venues + batch 3 retail
Art Museum	127	batch 3 arts
Exhibition	933	data/batch2-*.csv (104 centres + 797 halls)
Sports	338	batch 3 sports/racing + venue rows
POP-UP	54	batch 3 pop-up
CASES	1,309	data/cases.csv

Views overlap by design: All Venue contains every venue row; category views are subsets. Counts are per view and not additive.

Data rules

• Empty means not published, never zero. Nothing is interpolated.
• Coordinates are WGS-84 only.
• Status is open, closed or unknown, each with a source and an as-of date.
• Case records are links only — title, brand, year, publisher and the original URL.
• No personal contact data is published.

Loading

from datasets import load_dataset
hotels = load_dataset("eventthing/venue-library", "hotels", split="train")
venues = load_dataset("eventthing/venue-library", "venues", split="train")

Each CSV is exposed as its own config; manifest.json carries rows, columns and a SHA-256 per file.