Operationalizing a National Digital Library: The Case for a Norwegian Transformer Model
Paper • 2104.09617 • Published • 2
Token Classification
Transformers
TensorBoard
Safetensors
MLX
Norwegian
Norwegian Bokmål
Norwegian Nynorsk
bert
Trained with AutoTrain
Eval Results (legacy)
# Load model directly
from transformers import AutoTokenizer, AutoModelForTokenClassification
tokenizer = AutoTokenizer.from_pretrained("NbAiLab/nb-bert-base-pos")
model = AutoModelForTokenClassification.from_pretrained("NbAiLab/nb-bert-base-pos")Quick Links
Release 1.0 (November 6, 2024)
nb-bert-base-pos
Description
NB-Bert base model fine-tuned on the Part of Speech task using the NorNE dataset.
Usage
from transformers import pipeline
pos = pipeline("token-classification", "NbAiLab/nb-bert-base-pos")
example = "Jeg heter Kjell og bor i Oslo."
pos_results = pos(example)
print(ner_results)
More on https://arxiv.org/abs/2104.09617
- Downloads last month
- 63
Model size
0.2B params
Tensor type
F32
·
Hardware compatibility
Log In to add your hardware
Quantized
Model tree for NbAiLab/nb-bert-base-pos
Base model
NbAiLab/nb-bert-basePaper for NbAiLab/nb-bert-base-pos
Evaluation results
- Test Loss on ltgoslo/norneself-reported0.065
- Test Precision on ltgoslo/norneself-reported0.985
- Test Recall on ltgoslo/norneself-reported0.988
- Test F1 on ltgoslo/norneself-reported0.986
- Test Accuracy on ltgoslo/norneself-reported0.986
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="NbAiLab/nb-bert-base-pos")