iceman2434's picture
Create README.md
f3cf4bd verified
metadata
datasets:
  - universal_dependencies
language:
  - tl
metrics:
  - f1
pipeline_tag: token-classification

Model Specification

  • Model: XLM-RoBERTa (base-sized model)
  • Training Data:
    • Combined Afrikaans, Hebrew, Bulgarian, Vietnamese, Norwegian, Urdu, Czech, & Persian corpora (Top 8 Languages)
  • Training Details:
    • Base configurations with a minor adjustment in learning rate (4.5e-5)

Evaluation

  • Evaluation Dataset: Universal Dependencies Tagalog Ugnayan (Testing Set)
  • Tested in a zero-shot cross-lingual scenario on a Universal Dependencies Tagalog Ugnayan testing dataset (with 78.29% Accuracy)

POS Tags

  • ADJ – ADP – ADV – CCONJ – DET – INTJ – NOUN – NUM – PART – PRON – PROPN – PUNCT – SCONJ – VERB