Wildreceipts OCR Eval: Benchmark for Receipt Text Recognition

An evaluation dataset for optical character recognition (OCR) systems, focusing on receipts. The dataset was created by author mychen76 and was last updated on Hugging Face in September 2023. Full details are referenced in a related training dataset.

Use Cases

• Benchmarking OCR model accuracy based on receipt images.
• Training receipt-specific text recognition models based on the related training dataset.
• Evaluating model robustness on real-world, unstructured document layouts.

Strengths

• Dataset is specifically designed for OCR evaluation, providing a focused benchmark.
• Last update timestamp (2023-09-21 12:34:49) is explicitly provided.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Hugging Face (author: mychen76)

Freshness

Last updated 2023-09-21 12:34:49; freshness should be verified.