Hugging Face's logo

wietsedv
/
xlm-roberta-base-ft-udpos28-bg

Token Classification
Transformers
PyTorch
Safetensors
Bulgarian
xlm-roberta
part-of-speech
Eval Results
Inference Endpoints
Model card Files Community
1
Edit model card

XLM-RoBERTa base Universal Dependencies v2.8 POS tagging: Bulgarian

This model is part of our paper called:

  • Make the Best of Cross-lingual Transfer: Evidence from POS Tagging with over 100 Languages

Check the Space for more details.

Usage 

from transformers import AutoTokenizer, AutoModelForTokenClassification

tokenizer = AutoTokenizer.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-bg")
model = AutoModelForTokenClassification.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-bg")
Downloads last month
25
Safetensors
Model size
277M params
Tensor type
I64
·
F32
·

Dataset used to train wietsedv/xlm-roberta-base-ft-udpos28-bg

Space using wietsedv/xlm-roberta-base-ft-udpos28-bg 1

Evaluation results

  • English Test accuracy on Universal Dependencies v2.8
    self-reported
    85.000
  • Dutch Test accuracy on Universal Dependencies v2.8
    self-reported
    81.600
  • German Test accuracy on Universal Dependencies v2.8
    self-reported
    82.600
  • Italian Test accuracy on Universal Dependencies v2.8
    self-reported
    82.500
  • French Test accuracy on Universal Dependencies v2.8
    self-reported
    83.100
  • Spanish Test accuracy on Universal Dependencies v2.8
    self-reported
    85.500
  • Russian Test accuracy on Universal Dependencies v2.8
    self-reported
    92.700
  • Swedish Test accuracy on Universal Dependencies v2.8
    self-reported
    89.500
  • Norwegian Test accuracy on Universal Dependencies v2.8
    self-reported
    84.500
  • Danish Test accuracy on Universal Dependencies v2.8
    self-reported
    88.400
View on Papers With Code