Portuguese Handwritten Manuscripts from the 16th to 19th Centuries

16th to 19th century images and transcriptions of Portuguese manuscripts, primarily from Portuguese Inquisition court records. The dataset is designed for training and evaluating Handwritten Text Recognition (HTR) models and was authored by SSamDav on Hugging Face, last updated in October 2025.

Use Cases

• Train Handwritten Text Recognition (HTR) models based on historical Portuguese handwriting images and transcriptions.
• Benchmark OCR performance on cursive scripts from the 16th-19th centuries.
• Study linguistic and paleographic evolution in Portuguese court records over four centuries.

Strengths

• Covers a long time range from the 16th to the 19th century.
• Includes ground truth transcriptions for model training and evaluation.
• Focuses on a specific, historically significant source: Portuguese Inquisition court records.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• 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

Primarily derived from the court records of the Portuguese Inquisition.

Time Range

16th to 19th centuries

Freshness

Last updated 2025-10-13 19:58:33; freshness should be verified.