Dataset · yhcelebi
Synthetic Accounting and Operations of 30 Companies
yhcelebi/enterprise-data
A massive 9.3-billion-row tabular corpus with complete financial statements of fictional firms over 31 years.
What's inside
This is a fully synthetic and reconciled dataset of accounting and operational data for 30 fictional companies from January 1995 to September 2026. It contains approximately 9.3 billion rows in parquet format (about 48 GB) in English and Turkish: general ledgers, trial balances, accounts, payments, bank statements, loans, taxes, payrolls, and budgets.
What you can do with it
- Training and evaluating tabular models for time-series forecasting and regression
- Developing and testing Table QA systems for complex financial tables
- Tasks
- Tabular classification, Tabular regression, Time series forecasting, Table question answering
- Languages
- English, Turkish
- Size
- 1B–10B rows
- Downloads this month
- 45.3K
- Likes
- 2
How to load it
from datasets import load_dataset
ds = load_dataset("yhcelebi/enterprise-data")