Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: gpl-3.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: gpl-3.0
|
| 3 |
+
|
| 4 |
+
language:
|
| 5 |
+
- gl
|
| 6 |
+
- pt
|
| 7 |
+
|
| 8 |
+
widget:
|
| 9 |
+
|
| 10 |
+
- text: "A miña amiga Rosa, de Lisboa, estudou en Montreal. Agora traballa en Nova Pescanova."
|
| 11 |
+
|
| 12 |
---
|
| 13 |
+
|
| 14 |
+
# Named Entity Recognition (NER) model for Galician
|
| 15 |
+
|
| 16 |
+
This is a NER model for Galician (ILG/RAG spelling) which uses the standard `enamex' classes: LOC (geographical locations); PER (people); ORG (organizations); MISC (other entities).
|
| 17 |
+
|
| 18 |
+
The model is based on [BERT-base-gl-cased](https://huggingface.co/marcosgg/bert-base-gl-cased), which has been fine-tuned using custom splits of the [SLI_NERC dataset](https://github.com/xavier-gz/SLI_Galician_Corpora). On the test split of this dataset (not used for training), the model obtained the following results (Precision/Recall/F-score): 87.69 / 89.7 / 88.68
|