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

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

Downloads last month: 23

Safetensors

Model size

277M params

Tensor type

I64

F32

Inference Examples

Token Classification

Inference API (serverless) is not available, repository is disabled.

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

Spaces using wietsedv/xlm-roberta-base-ft-udpos28-de 2

Evaluation results

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

87.000
Dutch Test accuracy on Universal Dependencies v2.8
self-reported

89.600
German Test accuracy on Universal Dependencies v2.8
self-reported

97.200
Italian Test accuracy on Universal Dependencies v2.8
self-reported

85.600
French Test accuracy on Universal Dependencies v2.8
self-reported

84.800
Spanish Test accuracy on Universal Dependencies v2.8
self-reported

88.400
Russian Test accuracy on Universal Dependencies v2.8
self-reported

89.400
Swedish Test accuracy on Universal Dependencies v2.8
self-reported

92.300
Norwegian Test accuracy on Universal Dependencies v2.8
self-reported

87.700
Danish Test accuracy on Universal Dependencies v2.8
self-reported

88.900

View on Papers With Code