wietsedv's picture
update README
5b08566
metadata
language:
  - lv
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-lv
    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: 84.7
          - type: accuracy
            name: Dutch Test accuracy
            value: 85.6
          - type: accuracy
            name: German Test accuracy
            value: 82.5
          - type: accuracy
            name: Italian Test accuracy
            value: 84.3
          - type: accuracy
            name: French Test accuracy
            value: 84.1
          - type: accuracy
            name: Spanish Test accuracy
            value: 84.7
          - type: accuracy
            name: Russian Test accuracy
            value: 92.1
          - type: accuracy
            name: Swedish Test accuracy
            value: 86.8
          - type: accuracy
            name: Norwegian Test accuracy
            value: 81.3
          - type: accuracy
            name: Danish Test accuracy
            value: 86
          - type: accuracy
            name: Low Saxon Test accuracy
            value: 51.6
          - type: accuracy
            name: Akkadian Test accuracy
            value: 32.4
          - type: accuracy
            name: Armenian Test accuracy
            value: 87.5
          - type: accuracy
            name: Welsh Test accuracy
            value: 65.4
          - type: accuracy
            name: Old East Slavic Test accuracy
            value: 76.5
          - type: accuracy
            name: Albanian Test accuracy
            value: 75.9
          - type: accuracy
            name: Slovenian Test accuracy
            value: 82
          - type: accuracy
            name: Guajajara Test accuracy
            value: 31.1
          - type: accuracy
            name: Kurmanji Test accuracy
            value: 76.5
          - type: accuracy
            name: Turkish Test accuracy
            value: 77.2
          - type: accuracy
            name: Finnish Test accuracy
            value: 85.9
          - type: accuracy
            name: Indonesian Test accuracy
            value: 79.3
          - type: accuracy
            name: Ukrainian Test accuracy
            value: 91.1
          - type: accuracy
            name: Polish Test accuracy
            value: 88.5
          - type: accuracy
            name: Portuguese Test accuracy
            value: 84.9
          - type: accuracy
            name: Kazakh Test accuracy
            value: 83.8
          - type: accuracy
            name: Latin Test accuracy
            value: 81
          - type: accuracy
            name: Old French Test accuracy
            value: 56.7
          - type: accuracy
            name: Buryat Test accuracy
            value: 64.8
          - type: accuracy
            name: Kaapor Test accuracy
            value: 25
          - type: accuracy
            name: Korean Test accuracy
            value: 65.1
          - type: accuracy
            name: Estonian Test accuracy
            value: 84.7
          - type: accuracy
            name: Croatian Test accuracy
            value: 89.1
          - type: accuracy
            name: Gothic Test accuracy
            value: 23.5
          - type: accuracy
            name: Swiss German Test accuracy
            value: 45.2
          - type: accuracy
            name: Assyrian Test accuracy
            value: 12.8
          - type: accuracy
            name: North Sami Test accuracy
            value: 43.5
          - type: accuracy
            name: Naija Test accuracy
            value: 36.1
          - type: accuracy
            name: Latvian Test accuracy
            value: 96.9
          - type: accuracy
            name: Chinese Test accuracy
            value: 53.1
          - type: accuracy
            name: Tagalog Test accuracy
            value: 72.7
          - type: accuracy
            name: Bambara Test accuracy
            value: 28.6
          - type: accuracy
            name: Lithuanian Test accuracy
            value: 91
          - type: accuracy
            name: Galician Test accuracy
            value: 84.2
          - type: accuracy
            name: Vietnamese Test accuracy
            value: 65.7
          - type: accuracy
            name: Greek Test accuracy
            value: 84.5
          - type: accuracy
            name: Catalan Test accuracy
            value: 83.2
          - type: accuracy
            name: Czech Test accuracy
            value: 88
          - type: accuracy
            name: Erzya Test accuracy
            value: 52.5
          - type: accuracy
            name: Bhojpuri Test accuracy
            value: 49.2
          - type: accuracy
            name: Thai Test accuracy
            value: 63.3
          - type: accuracy
            name: Marathi Test accuracy
            value: 85.3
          - type: accuracy
            name: Basque Test accuracy
            value: 77.4
          - type: accuracy
            name: Slovak Test accuracy
            value: 87.8
          - type: accuracy
            name: Kiche Test accuracy
            value: 40.3
          - type: accuracy
            name: Yoruba Test accuracy
            value: 28.4
          - type: accuracy
            name: Warlpiri Test accuracy
            value: 44.9
          - type: accuracy
            name: Tamil Test accuracy
            value: 86.4
          - type: accuracy
            name: Maltese Test accuracy
            value: 25.9
          - type: accuracy
            name: Ancient Greek Test accuracy
            value: 62.2
          - type: accuracy
            name: Icelandic Test accuracy
            value: 81.7
          - type: accuracy
            name: Mbya Guarani Test accuracy
            value: 35.3
          - type: accuracy
            name: Urdu Test accuracy
            value: 61.9
          - type: accuracy
            name: Romanian Test accuracy
            value: 82.2
          - type: accuracy
            name: Persian Test accuracy
            value: 74.8
          - type: accuracy
            name: Apurina Test accuracy
            value: 49
          - type: accuracy
            name: Japanese Test accuracy
            value: 39.4
          - type: accuracy
            name: Hungarian Test accuracy
            value: 79.9
          - type: accuracy
            name: Hindi Test accuracy
            value: 64.1
          - type: accuracy
            name: Classical Chinese Test accuracy
            value: 30
          - type: accuracy
            name: Komi Permyak Test accuracy
            value: 51.7
          - type: accuracy
            name: Faroese Test accuracy
            value: 76.2
          - type: accuracy
            name: Sanskrit Test accuracy
            value: 39.7
          - type: accuracy
            name: Livvi Test accuracy
            value: 67.7
          - type: accuracy
            name: Arabic Test accuracy
            value: 79.4
          - type: accuracy
            name: Wolof Test accuracy
            value: 31.7
          - type: accuracy
            name: Bulgarian Test accuracy
            value: 89
          - type: accuracy
            name: Akuntsu Test accuracy
            value: 35.5
          - type: accuracy
            name: Makurap Test accuracy
            value: 20.5
          - type: accuracy
            name: Kangri Test accuracy
            value: 50.6
          - type: accuracy
            name: Breton Test accuracy
            value: 62.7
          - type: accuracy
            name: Telugu Test accuracy
            value: 87.8
          - type: accuracy
            name: Cantonese Test accuracy
            value: 50.7
          - type: accuracy
            name: Old Church Slavonic Test accuracy
            value: 49.3
          - type: accuracy
            name: Karelian Test accuracy
            value: 72.7
          - type: accuracy
            name: Upper Sorbian Test accuracy
            value: 75.6
          - type: accuracy
            name: South Levantine Arabic Test accuracy
            value: 68.7
          - type: accuracy
            name: Komi Zyrian Test accuracy
            value: 44.5
          - type: accuracy
            name: Irish Test accuracy
            value: 64.7
          - type: accuracy
            name: Nayini Test accuracy
            value: 39.7
          - type: accuracy
            name: Munduruku Test accuracy
            value: 26
          - type: accuracy
            name: Manx Test accuracy
            value: 37.9
          - type: accuracy
            name: Skolt Sami Test accuracy
            value: 34.7
          - type: accuracy
            name: Afrikaans Test accuracy
            value: 81.6
          - type: accuracy
            name: Old Turkish Test accuracy
            value: 22.6
          - type: accuracy
            name: Tupinamba Test accuracy
            value: 40.6
          - type: accuracy
            name: Belarusian Test accuracy
            value: 91.8
          - type: accuracy
            name: Serbian Test accuracy
            value: 89.7
          - type: accuracy
            name: Moksha Test accuracy
            value: 48.7
          - type: accuracy
            name: Western Armenian Test accuracy
            value: 77.5
          - type: accuracy
            name: Scottish Gaelic Test accuracy
            value: 58.1
          - type: accuracy
            name: Khunsari Test accuracy
            value: 40.5
          - type: accuracy
            name: Hebrew Test accuracy
            value: 85.4
          - type: accuracy
            name: Uyghur Test accuracy
            value: 79.7
          - type: accuracy
            name: Chukchi Test accuracy
            value: 37

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

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