metadata
language: fa
pipeline_tag: token-classification
library_name: transformers
QomSSLab/Verdict_Splitter
This repository hosts an XLM-RoBERTa token-classification head trained.
Usage
from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
model_id = "QomSSLab/Verdict_Splitter"
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
Oاستدلالتصمیمخارجخلعمقدمهپایانی
Metrics
Validation Metrics
- Precision: 0.7067
- Recall: 0.8457
- F1: 0.7700
- Accuracy: 0.9730
Per-label Breakdown
| Label | Precision | Recall | F1 | Support |
|---|---|---|---|---|
| O | 0.8617 | 0.8223 | 0.8416 | 394 |
| استدلال | 0.9733 | 0.9394 | 0.9561 | 6635 |
| تصمیم | 0.9895 | 0.9700 | 0.9797 | 5361 |
| خارج | 1.0000 | 1.0000 | 1.0000 | 0 |
| خلع | 1.0000 | 1.0000 | 1.0000 | 0 |
| مقدمه | 0.9689 | 0.9981 | 0.9833 | 10871 |
| پایانی | 0.9722 | 0.9879 | 0.9800 | 1732 |