metadata
language:
- ja
license: apache-2.0
library_name: transformers
tags:
- part-of-speech
- token-classification
datasets:
- universal_dependencies
metrics:
- accuracy
model-index:
- name: xlm-roberta-base-ft-udpos28-ja
results:
- task:
type: token-classification
name: Part-of-Speech Tagging
dataset:
type: universal_dependencies
name: Universal Dependencies v2.8
metrics:
- type: accuracy
name: English Test accuracy
value: 47.7
- type: accuracy
name: Dutch Test accuracy
value: 49.8
- type: accuracy
name: German Test accuracy
value: 55.7
- type: accuracy
name: Italian Test accuracy
value: 52
- type: accuracy
name: French Test accuracy
value: 47.2
- type: accuracy
name: Spanish Test accuracy
value: 48.2
- type: accuracy
name: Russian Test accuracy
value: 62.7
- type: accuracy
name: Swedish Test accuracy
value: 52.6
- type: accuracy
name: Norwegian Test accuracy
value: 48.6
- type: accuracy
name: Danish Test accuracy
value: 54.3
- type: accuracy
name: Low Saxon Test accuracy
value: 34.7
- type: accuracy
name: Akkadian Test accuracy
value: 38.6
- type: accuracy
name: Armenian Test accuracy
value: 67
- type: accuracy
name: Welsh Test accuracy
value: 48.4
- type: accuracy
name: Old East Slavic Test accuracy
value: 55.2
- type: accuracy
name: Albanian Test accuracy
value: 51.8
- type: accuracy
name: Slovenian Test accuracy
value: 46.6
- type: accuracy
name: Guajajara Test accuracy
value: 39.3
- type: accuracy
name: Kurmanji Test accuracy
value: 54.6
- type: accuracy
name: Turkish Test accuracy
value: 65.4
- type: accuracy
name: Finnish Test accuracy
value: 69.1
- type: accuracy
name: Indonesian Test accuracy
value: 59.1
- type: accuracy
name: Ukrainian Test accuracy
value: 63.2
- type: accuracy
name: Polish Test accuracy
value: 60.5
- type: accuracy
name: Portuguese Test accuracy
value: 53.3
- type: accuracy
name: Kazakh Test accuracy
value: 71.9
- type: accuracy
name: Latin Test accuracy
value: 53.5
- type: accuracy
name: Old French Test accuracy
value: 30
- type: accuracy
name: Buryat Test accuracy
value: 58.2
- type: accuracy
name: Kaapor Test accuracy
value: 21.7
- type: accuracy
name: Korean Test accuracy
value: 64.5
- type: accuracy
name: Estonian Test accuracy
value: 67
- type: accuracy
name: Croatian Test accuracy
value: 57.5
- type: accuracy
name: Gothic Test accuracy
value: 15.4
- type: accuracy
name: Swiss German Test accuracy
value: 34.5
- type: accuracy
name: Assyrian Test accuracy
value: 28.3
- type: accuracy
name: North Sami Test accuracy
value: 35.1
- type: accuracy
name: Naija Test accuracy
value: 16.8
- type: accuracy
name: Latvian Test accuracy
value: 69.6
- type: accuracy
name: Chinese Test accuracy
value: 66.2
- type: accuracy
name: Tagalog Test accuracy
value: 50.4
- type: accuracy
name: Bambara Test accuracy
value: 27.5
- type: accuracy
name: Lithuanian Test accuracy
value: 69.7
- type: accuracy
name: Galician Test accuracy
value: 51.6
- type: accuracy
name: Vietnamese Test accuracy
value: 50.6
- type: accuracy
name: Greek Test accuracy
value: 54.9
- type: accuracy
name: Catalan Test accuracy
value: 46.1
- type: accuracy
name: Czech Test accuracy
value: 61.1
- type: accuracy
name: Erzya Test accuracy
value: 41.3
- type: accuracy
name: Bhojpuri Test accuracy
value: 41.9
- type: accuracy
name: Thai Test accuracy
value: 52.3
- type: accuracy
name: Marathi Test accuracy
value: 77.3
- type: accuracy
name: Basque Test accuracy
value: 68.4
- type: accuracy
name: Slovak Test accuracy
value: 62.3
- type: accuracy
name: Kiche Test accuracy
value: 41
- type: accuracy
name: Yoruba Test accuracy
value: 28.8
- type: accuracy
name: Warlpiri Test accuracy
value: 30.4
- type: accuracy
name: Tamil Test accuracy
value: 75.9
- type: accuracy
name: Maltese Test accuracy
value: 29.8
- type: accuracy
name: Ancient Greek Test accuracy
value: 50.2
- type: accuracy
name: Icelandic Test accuracy
value: 54.4
- type: accuracy
name: Mbya Guarani Test accuracy
value: 28.1
- type: accuracy
name: Urdu Test accuracy
value: 46.4
- type: accuracy
name: Romanian Test accuracy
value: 55.4
- type: accuracy
name: Persian Test accuracy
value: 51.8
- type: accuracy
name: Apurina Test accuracy
value: 34.5
- type: accuracy
name: Japanese Test accuracy
value: 92.6
- type: accuracy
name: Hungarian Test accuracy
value: 61.2
- type: accuracy
name: Hindi Test accuracy
value: 48.2
- type: accuracy
name: Classical Chinese Test accuracy
value: 46.1
- type: accuracy
name: Komi Permyak Test accuracy
value: 42.8
- type: accuracy
name: Faroese Test accuracy
value: 51.1
- type: accuracy
name: Sanskrit Test accuracy
value: 33
- type: accuracy
name: Livvi Test accuracy
value: 57.2
- type: accuracy
name: Arabic Test accuracy
value: 52.7
- type: accuracy
name: Wolof Test accuracy
value: 32.1
- type: accuracy
name: Bulgarian Test accuracy
value: 55.1
- type: accuracy
name: Akuntsu Test accuracy
value: 41.4
- type: accuracy
name: Makurap Test accuracy
value: 19.9
- type: accuracy
name: Kangri Test accuracy
value: 41
- type: accuracy
name: Breton Test accuracy
value: 46.4
- type: accuracy
name: Telugu Test accuracy
value: 71.8
- type: accuracy
name: Cantonese Test accuracy
value: 60.4
- type: accuracy
name: Old Church Slavonic Test accuracy
value: 39.5
- type: accuracy
name: Karelian Test accuracy
value: 60.7
- type: accuracy
name: Upper Sorbian Test accuracy
value: 54.6
- type: accuracy
name: South Levantine Arabic Test accuracy
value: 49.4
- type: accuracy
name: Komi Zyrian Test accuracy
value: 39.8
- type: accuracy
name: Irish Test accuracy
value: 46.8
- type: accuracy
name: Nayini Test accuracy
value: 37.2
- type: accuracy
name: Munduruku Test accuracy
value: 39.3
- type: accuracy
name: Manx Test accuracy
value: 33.9
- type: accuracy
name: Skolt Sami Test accuracy
value: 36.4
- type: accuracy
name: Afrikaans Test accuracy
value: 45.7
- type: accuracy
name: Old Turkish Test accuracy
value: 18.1
- type: accuracy
name: Tupinamba Test accuracy
value: 32
- type: accuracy
name: Belarusian Test accuracy
value: 62.6
- type: accuracy
name: Serbian Test accuracy
value: 58
- type: accuracy
name: Moksha Test accuracy
value: 42.2
- type: accuracy
name: Western Armenian Test accuracy
value: 62.3
- type: accuracy
name: Scottish Gaelic Test accuracy
value: 38.6
- type: accuracy
name: Khunsari Test accuracy
value: 44.6
- type: accuracy
name: Hebrew Test accuracy
value: 69.8
- type: accuracy
name: Uyghur Test accuracy
value: 65.4
- type: accuracy
name: Chukchi Test accuracy
value: 33.7
XLM-RoBERTa base Universal Dependencies v2.8 POS tagging: Japanese
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-ja")
model = AutoModelForTokenClassification.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-ja")