QomSSLab/etghan_tagger_v1

This repository hosts an XLM-RoBERTa token-classification head trained.

Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline

model_id = "QomSSLab/etghan_tagger_v1"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForTokenClassification.from_pretrained(model_id)
tagger = pipeline("token-classification", model=model, tokenizer=tokenizer, aggregation_strategy="simple")

text = "مثال از یک ورودی فارسی"
for entity in tagger(text):
    print(entity)

Labels

• A
• ABSENCE
• APPEALABILITY
• APPEAL_AUTHORITY
• APPEAL_DEADLINE
• B
• BSEMHE
• COURT_NAME
• DEFENDANT_ADDRESS
• DEFENDANT_FATHER
• DEFENDANT_NAME
• FINALITY
• JUDGE_NAME
• JUDGE_POSITION
• JUDGE_SIGNATURE
• JUDGMENT_DATE
• LEGAL_CHARGE
• LEGAL_REPRESENTATIVE
• O
• PLACE_OF_OFFENSE
• PLACE_OF_OFFICE
• PLAINTIFF_ADDRESS
• PLAINTIFF_CLAIM
• PLAINTIFF_FATHER
• PLAINTIFF_NAME
• PRESENCE
• SUBJECT_OF_CLAIM
• TIME_OF_OFFENSE

Training

• Base model: xlm-roberta-large
• Optimizer/args: default Trainer settings (AdamW, lr=3e-5, batch size 8, epochs configurable)