| Feature | Description |
|---|---|
| Name | de_APIS_OEBL_NER_lg |
| Version | 1.0 |
| spaCy | >=3.6.0,<3.7.0 |
| Default Pipeline | tok2vec, ner |
| Components | tok2vec, ner |
| Vectors | 500000 keys, 500000 unique vectors (300 dimensions) |
| Sources | n/a |
| License | n/a |
| Author | n/a |
Label Scheme
View label scheme (3 labels for 1 components)
| Component | Labels |
|---|---|
ner |
LOC, ORG, PER |
Accuracy
| Type | Score |
|---|---|
ENTS_F |
77.85 |
ENTS_P |
76.71 |
ENTS_R |
79.03 |
TOK2VEC_LOSS |
13266.36 |
NER_LOSS |
378634.81 |
Sources
Trained on data originating from the APIS project and the Austrian Biographical Lexicon (ÖBL).
Reproducible training context (model m2): https://github.com/acdh-oeaw/veld_chain_7_train/
Dataset available here: https://huggingface.co/datasets/SteffRhes/APIS_OEBL__Named_Entity_Recognition
- Downloads last month
- 3
Dataset used to train SteffRhes/de_APIS_OEBL_NER_lg
Evaluation results
- NER Precisionself-reported0.767
- NER Recallself-reported0.790
- NER F Scoreself-reported0.779