metadata
language:
- ar
datasets:
- AJGT
tags:
- AJGT
widget:
- text: يهدي الله من يشاء
- text: الاسلوب قذر وقمامه
BERT-AJGT
Arabic version bert model fine tuned on AJGT dataset
Data
The model were fine-tuned on ~1800 sentence from twitter for Jordanian dialect.
Results
| class | precision | recall | f1-score | Support |
|---|---|---|---|---|
| 0 | 0.9462 | 0.9778 | 0.9617 | 90 |
| 1 | 0.9399 | 0.9689 | 0.9542 | 90 |
| Accuracy | 0.9611 | 180 |
How to use
You can use these models by installing torch or tensorflow and Huggingface library transformers. And you can use it directly by initializing it like this:
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model_name="mofawzy/bert-ajgt"
model = AutoModelForSequenceClassification.from_pretrained(model_name,num_labels=2)
tokenizer = AutoTokenizer.from_pretrained(model_name)