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

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-sa")
model = AutoModelForTokenClassification.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-sa")

Downloads last month: 1

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

Evaluation results

English Test accuracy on Universal Dependencies v2.8
self-reported

31.400
Dutch Test accuracy on Universal Dependencies v2.8
self-reported

28.400
German Test accuracy on Universal Dependencies v2.8
self-reported

32.300
Italian Test accuracy on Universal Dependencies v2.8
self-reported

28.300
French Test accuracy on Universal Dependencies v2.8
self-reported

28.100
Spanish Test accuracy on Universal Dependencies v2.8
self-reported

28.500
Russian Test accuracy on Universal Dependencies v2.8
self-reported

37.500
Swedish Test accuracy on Universal Dependencies v2.8
self-reported

35.700