app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# موديل BRET
model_name = "NvbilVmir1/Madar"

# تحميل الموديل والتوكنيزر
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
id2label = model.config.id2label

def predict_dialect(text):
    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
    with torch.no_grad():
        outputs = model(**inputs)
        pred = torch.argmax(outputs.logits, dim=-1).item()
    return id2label[pred]

title = "AraELECTRA Arabic Dialect Classifier 🌍"
description = """
Detect the Arabic dialect from the input text.
Dialects include: African, Khaleeji, Levant, and Egyptian.
"""

interface = gr.Interface(
    fn=predict_dialect,
    inputs=gr.Textbox(lines=3, placeholder="Type an Arabic sentence here..."),
    outputs="text",
    title=title,
    description=description,
    theme="default",
    allow_flagging="never",
)

if __name__ == "__main__":
    interface.launch()