Persian OCR 230K: A Dataset for Optical Character Recognition in Persian

230,000 total samples for Persian optical character recognition, split into 184,000 training and 46,000 test examples. The dataset was uploaded by user 'ordaktaktak' to Hugging Face and was last updated on September 26, 2024. It likely contains images of Persian text paired with corresponding transcriptions.

Use Cases

• Train optical character recognition models based on the described image-text pairs.
• Benchmark the performance of multilingual OCR systems on Persian script.
• Fine-tune pre-trained vision-language models for Persian text extraction tasks.

Strengths

• Provides a substantial scale with 230,000 total samples.
• Includes a predefined split of 184,000 training and 46,000 test examples.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Hugging Face user 'ordaktaktak'.

Freshness

Last updated 2024-09-26 06:55:50; freshness should be verified.