Skip to content

Spans (NER) format

🚧 In progress

This workflow is implemented but has not yet been run end-to-end, so this page is a stub. Detailed usage and results will land here once the pipeline is verified. Until then, treat the entry point and config below as the source of truth.

to_spans.py converts every ner/* entry in the task registry into GLiNER-style span training data, reading the extracted text plus its annotations sidecar (which must provide a spans field). It emits tasks/ner/<dataset>/<split>.jsonl.gz. The legacy --schema {gliner|conll|generic} flag is gone — only the GLiNER schema is produced.

Entry point: scripts/data/to_spans.py Config: configs/data/tasks.yaml