Barbados historical handwriting review checkpoints

This repository supplies trained model artifacts for the reviewed historical handwriting recognition solution. The code package is delivered separately to the organizers. It contains no competition images, label tables, OCR predictions or submission CSVs.

Reported public score: 0.927525446. WER Weighted: 1.306287832. CER Weighted: 1.985048359. Private score: not supplied.

manifest.json maps every checkpoint and fitted selector to a SHA256-addressed file under weights/. Identical files share one stored object. Restore each target_path beneath the configured trained-model directory and link it into the runtime workspace. The organizer code package pins a completed commit.

Public pretrained initialization models are downloaded from their upstream repositories and are not duplicated here. Model families include TrOCR line/word/phrase OCR, Kraken CTC, Qwen OCR LoRA variants and CPU selectors. The inventory retains historical initialization/support artifacts conservatively and is not claimed to be the smallest possible model set.

Upstream model licenses and the competition's data-use terms continue to apply. This card does not grant rights to the competition dataset or claim that the complete raw-image pipeline has passed a fresh clean-environment run. Consult the organizer code package for training, inference, runtime and validation scope.