A dataset of PDF documents annotated for OCR classification tasks, published by HuggingFaceFW. It contains binary labels (OCR/NOCR) and file size information for each PDF. The dataset was last updated on October 20, 2025.
Use Cases
- Training binary classifiers to detect OCR-processed PDFs based on the 'class' label.
- Analyzing the relationship between PDF file size and OCR status based on the 'pdf_size_bytes' column.
- Evaluating model performance on truncated versus non-truncated PDFs based on the 'truncation_type' feature.
- Benchmarking OCR detection algorithms using a dataset with a known class distribution (1393 NOCR, 227 OCR).
Strengths
- Provides explicit binary classification labels ('OCR'/'NOCR') for each PDF.
- Includes a specific class distribution: 1393 NOCR samples and 227 OCR samples.
- Contains metadata such as file size ('pdf_size_bytes') and truncation status ('truncation_type').
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Row count is unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
Provenance
- Source
- HuggingFaceFW
- Freshness
- Last updated 2025-10-20 17:29:22; freshness should be verified.