Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
accuracy
float64
macro_f1
float64
sample_only_accuracy
float64
passed
bool
1
1
1
true

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

MailRoom Document Corpus

Synthetic enterprise mailroom documents for classification training and evaluation.

Contents

File Records Description
documents.jsonl 48 12 base samples × 4 variants (augmented)
manifest.json Dataset metadata
benchmark_report.json Classifier evaluation metrics

Categories

  • leave_request — HR leave/time-off forms
  • correspondence — Official letters and memos
  • invoice — Tax invoices and financial documents
  • complaint — Customer grievances
  • purchase_request — RFQ and procurement requests
  • hr_inquiry — Policy and benefits questions

Languages

English (en) and Persian (fa) — bilingual registry workflow.

Usage

import json
with open("documents.jsonl") as f:
    for line in f:
        doc = json.loads(line)
        print(doc["category"], doc["filename"])

License

Apache 2.0 · Synthetic data — no real PII.

Downloads last month
26