AyoubChLin/Company-document-dataset-v2

Company Documents v2

Generation complete: all 13 document types have completed export and upload checkpoints.

Synthetic, born-digital business documents rendered from four open sample databases, with exact gold labels: 353,580 PDFs (404,514 pages) of 13 document types in English and French, issued by 60 synthetic companies, each with its own letterhead, numbering and wording. Successor of CompanyDocuments (2,677 PDFs, 4 types).

Dataset overview

property	value
PDF documents	353,580
PDF pages	404,514
Document types	13
Synthetic issuing companies	60
Languages	English / French
Degraded scan/photo page images	120,523
Export format	parquet
Split unit	issuing company, grouped by sector

Intended uses

Document classification, OCR evaluation, key information extraction, layout-aware field extraction, and document question answering using the gold fields to construct task-specific examples. The dataset provides document text, structured fields and coordinates; it does not include a separate set of question-answer pairs.

Subsets

Pick a document type, or all. Every subset has the same train / validation / test split.

subset	document	documents	train	validation	test	sources
all (default)	every type below	353,580	259,497	47,114	46,969	all four
account_statement	a customer's invoices and payments over a period, running balance and aging	728	522	99	107	adventureworks, northwind
credit_note	credit against an invoice, with a reason (damaged, wrong item, price...)	47,747	35,107	6,337	6,303	adventureworks, northwind
employment_certificate	letter certifying an employee's job title, department and hire date	290	202	43	45	adventureworks
goods_received_note	delivery check against a purchase order: ordered, received, accepted, rejected	3,689	2,667	512	510	adventureworks
inventory_report	stock on hand per product, with value and reorder status	51	35	6	10	adventureworks, northwind
invoice	sales invoice: line items, discounts, tax, totals, payment terms	48,158	35,342	6,446	6,370	adventureworks, chinook, northwind
packing_slip	items packed for a shipment: ordered vs shipped quantities	47,740	35,040	6,456	6,244	adventureworks, northwind
payslip	employee pay for a period: earnings, deductions, net pay	290	231	30	29	adventureworks
purchase_order	order the issuer places with a vendor	50,313	36,908	6,598	6,807	adventureworks, northwind
quote	price quotation with validity and lead time	47,746	35,004	6,313	6,429	adventureworks, northwind
receipt	proof of a paid retail transaction (A5)	16,456	12,105	2,218	2,133	chinook, sakila
shipping_order	instruction to ship an order: items, weights, consignee, carrier	47,747	35,182	6,274	6,291	adventureworks, northwind
work_order	production order: routing operations, planned vs actual cost	42,625	31,152	5,782	5,691	adventureworks
scans	degraded page images of a sample of the documents	120,523 pages	88,799	15,829	15,895

Quick start

import json
from datasets import load_dataset

documents = load_dataset("AyoubChLin/Company-document-dataset-v2", "invoice", split="train")
row = documents[0]
row["pdf"]                            # pdfplumber.PDF
json.loads(row["extracted_data"])     # the gold fields

Use "all" to load every document type. For a large subset, pass streaming=True to load_dataset and read examples with next(iter(documents)) to avoid downloading the entire subset.

Scans

120,523 degraded page images (scan and photo profiles) of a sample of the documents, with the gold-field boxes moved into image pixels:

scans = load_dataset("AyoubChLin/Company-document-dataset-v2", "scans", split="test")
scans[0]["image"], json.loads(scans[0]["boxes"])

Files

data/<type>/<split>-NNNNN.parquet    one row per PDF, the PDF embedded (the columns below)
scans/<type>/<split>-NNNNN.parquet   one row per degraded page image
stats.json                           counts per split, type, source, locale, layout; companies per split
licenses/                            source database licenses

Splits

Splits are by issuing company (per sector): validation and test documents come from companies whose issuing company is absent from training. Source databases, products, counterparties, layout families and themes can occur across splits; this is an issuer-held-out split, not a source-held-out split.

split	documents	companies	scan pages
train	259,497	44	88,799
validation	47,114	8	15,829
test	46,969	8	15,895

Columns

One row per PDF (all and every document-type subset):

column	content
pdf	the PDF itself (the viewer shows its first page; datasets opens it with pdfplumber; raw bytes with .cast_column("pdf", Pdf(decode=False)))
file_name	<type>/<doc_id>.pdf
doc_id, split, document_type	identifiers
source, variant	source database and document variant (<type>.<source>)
company, company_name, company_sector, company_country, company_city, company_tax_id, company_locale, company_currency	the issuing company
number, issue_date, currency, total, items_count, counterparty_role, counterparty_name	key gold values as plain columns
layout, theme, locale, pages	letterhead layout, color theme, language, page count
extracted_data	all gold fields as JSON
file_content	text extracted from the PDF, in reading order
options	content variation drawn for the document (JSON)
rendered	each gold field as printed in the PDF (JSON)
boxes	where each gold field is printed (JSON: field path -> list of page + bbox in PDF points)
words	every word of the PDF with its page and bbox (JSON)

scans subset, one row per degraded page image (profiles scan and photo: skew, blur, noise, paper tint, lighting, JPEG):

column	content
image	the page image
doc_id, split, document_type, page, profile	identifiers; join to the documents on doc_id
width, height	image size in pixels
params	degradation parameters (JSON)
words, boxes	words and gold-field boxes moved into image pixels (JSON)

Sources

source	documents
adventureworks	208,870
northwind	127,842
sakila	16,044
chinook	824

Northwind (food wholesale), AdventureWorks (bicycle manufacturer: sales, purchasing, production, HR), Chinook (online music store), Sakila (video rental, dates shifted +19 years). Issuers are synthetic; counterparties, products, quantities and prices come from the databases.

language	documents
English	203,657
French	149,923

Guarantees

Every document passed a self-check: the checked gold values are present in the PDF text and the arithmetic holds (line totals, tax, totals, running balances, aging, payroll, received vs accepted). Money is computed with Decimal and half-up rounding; currencies are converted once from the source (USD).

Limitations

• These are synthetic documents derived from sample databases. Their layouts, wording and simulated degradation do not cover the full variety of real business documents or camera captures.
• Document types are imbalanced; use the per-type counts when selecting training data and reporting evaluation results. Some types contain only a small number of examples.
• Scan rows represent pages of sampled PDFs, not additional independent business transactions. Keep scans with their parent document's split when constructing an evaluation dataset.
• Gold fields and word boxes come from the generation and PDF extraction pipeline. Arithmetic and text checks do not establish that every annotation or reading-order decision is error-free.
• The same underlying source records can support multiple document types. Holding out issuing companies does not guarantee that all business entities or transaction content are unseen.

Licenses

Code and generated documents: Apache-2.0. Source data: Northwind (MIT), AdventureWorks (MIT), Chinook (MIT), Sakila (BSD-2); see licenses/.

Citation

Created by Cherguelaine Ayoub. If you use this dataset, please cite:

@misc{cherguelaine2026companydocumentsv2,
  author = {Cherguelaine, Ayoub},
  title = {Company Documents v2: Synthetic Business Documents},
  year = {2026},
  url = {https://github.com/AyoubCherguelaine/Company-document-dataset-v2}
}

Reproduce

Generator source code: Company-document-dataset-v2. To generate the content locally from a Git checkout:

git clone https://github.com/AyoubCherguelaine/Company-document-dataset-v2.git
cd Company-document-dataset-v2
# Generate all source records and their scans locally.
python3 scripts/run_pipeline.py --all-records --stages venv download index companies texts generate augment --seed 42 --workers 8 --augment-fraction 0.3
# Export with the published dataset's company split seed.
./venv/bin/python -m docgen export output/v2 dataset/full-local --format parquet --seed 0

See the repository README for setup, configuration and full-dataset generation options.

Generated with the CompanyDocuments v2 generator (docgen) and this run config:

{
  "run_name": "full",
  "seed": 42,
  "out_dir": "output/parts/<type>",
  "workers": 8,
  "fx": {
    "USD": "1",
    "EUR": "0.92",
    "GBP": "0.79"
  },
  "self_check": true,
  "boxes": true,
  "resume": true,
  "documents": [
    {
      "type": "account_statement",
      "count": "all"
    },
    {
      "type": "credit_note",
      "count": "all"
    },
    {
      "type": "employment_certificate",
      "count": "all"
    },
    {
      "type": "goods_received_note",
      "count": "all"
    },
    {
      "type": "inventory_report",
      "count": "all"
    },
    {
      "type": "invoice",
      "count": "all"
    },
    {
      "type": "packing_slip",
      "count": "all"
    },
    {
      "type": "payslip",
      "count": "all"
    },
    {
      "type": "purchase_order",
      "count": "all"
    },
    {
      "type": "quote",
      "count": "all"
    },
    {
      "type": "receipt",
      "count": "all"
    },
    {
      "type": "shipping_order",
      "count": "all"
    },
    {
      "type": "work_order",
      "count": "all"
    }
  ]
}