aboutcode-org / aboutcode-org/scancode-toolkit
[GSoC] Train DeBERTa NER model for required phrase prediction
未關閉
GSoC
- 主要語言
- Python
- 星號
- 2.6k
- 分支
- 791
- 平均合併
- 1 天 12 小時
- 30 天內合併 PR
- 5
描述
continuation of #5077. uses the BIOES dataset produced there
## Goal
Finetune DeBERTa-v3-large to predict required phrase spans in license rule text using BIOES labels
## Tasks
- Subword tokenization with label alignment (-100 for continuation tokens)
- Training with class-weighted loss (handle O vs B/I/E/S imbalance)
- Include negative samples (rules without markers, all-O labels) for balanced training
- Evaluation: token F1, exact span match
- ONNX export for CPU inference
貢獻指南
評估
這個 Issue 還沒有評估資料。