|
import gradio as gr |
|
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline |
|
|
|
model_id = "xlm-roberta-base" |
|
peft_model_id = "rasyosef/xlm-roberta-base-lora-amharic-news-classification" |
|
|
|
categories = ['แแแญ แ แแ แแ', 'แแแแ', 'แตแแญแต', 'แขแแแต', 'แแแ แ แแ แแ', 'แแแฒแซ'] |
|
id2label = {i: lbl for i, lbl in enumerate(categories)} |
|
label2id = {lbl: i for i, lbl in enumerate(categories)} |
|
|
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
|
|
model = AutoModelForSequenceClassification.from_pretrained( |
|
model_id, |
|
num_labels=len(categories), |
|
id2label=id2label, |
|
label2id=label2id |
|
) |
|
|
|
model.load_adapter(peft_model_id) |
|
|
|
classifier = pipeline("text-classification", model=model, tokenizer=tokenizer) |
|
|
|
def predict(text): |
|
return classifier([text])[0] |
|
|
|
|
|
with gr.Blocks() as demo: |
|
gr.Markdown( |
|
""" |
|
# Amharic News Article Classification |
|
This RoBERTa model (xlm-roberta-base) was finetuned using Low-Rank Adaptation (LoRA) that classifies amharic news articles into one of the following 6 categories. |
|
- แแแญ แ แแ แแ (Local News) |
|
- แแแแ (Entertainment) |
|
- แตแแญแต (Sports) |
|
- แขแแแต (Business) |
|
- แแแ แ แแ แแ (International News) |
|
- แแแฒแซ (Politics) |
|
""" |
|
) |
|
|
|
with gr.Row(): |
|
with gr.Column(): |
|
input = gr.Textbox(label="Amharic text", placeholder="Enter text here", lines=3) |
|
classify_btn = gr.Button(value="Classify") |
|
with gr.Column(): |
|
output = gr.Textbox(label="Predicted class") |
|
|
|
classify_btn.click(predict, inputs=input, outputs=output) |
|
examples = gr.Examples( |
|
examples=[ |
|
"แขแตแฎแตแซ แแชแแจแญ แแ 6แ แณแแแต แจแฅแแต แจแแณแแฝ แ
แตแ แณแฐแณ", |
|
"แ แ แ แดแแตแฎแต แจแแแตแ แฆแณ แฐแจแตแ แแชแซแ แฐแแแฎ แจแแแ แจแแแจแ แแแฃแณ แฐแ แแแ แตแซ แฅแแฒแแแญ แแแชแแฝ แ แญแแแแข" |
|
], |
|
inputs=[input], |
|
) |
|
|
|
demo.launch() |
|
|
