app.py

import gradio as gr
from sentence_transformers import SentenceTransformer
import numpy as np

# Load model once at startup
st_model = SentenceTransformer("sentence-transformers/all-mpnet-base-v2")

TITLE = "# Text → Vector (all-mpnet-base-v2)"
DESC = (
    "Masukkan **kalimat** lalu dapatkan **embedding vector** "
    "(opsional dinormalisasi L2). Model: `sentence-transformers/all-mpnet-base-v2`."
)

def embed(text: str, normalize: bool = True):
    text = (text or "").strip()
    if not text:
        return [], 0
    vec = st_model.encode([text], normalize_embeddings=normalize)[0]
    return vec.tolist(), int(vec.shape[0])

with gr.Blocks() as demo:
    gr.Markdown(TITLE)
    gr.Markdown(DESC)

    with gr.Row():
        text_in = gr.Textbox(
            label="Kalimat",
            placeholder="Tulis kalimat di sini...",
            lines=3,
        )
    normalize = gr.Checkbox(value=True, label="Normalize embedding (L2)")
    btn = gr.Button("Compute Embedding", variant="primary")

    with gr.Row():
        vec_out = gr.JSON(label="Vector (list of floats)")
        dim_out = gr.Number(label="Dimensi vektor", interactive=False)

    gr.Examples(
        examples=[
            ["Halo dunia!"],
            ["Machine learning is fun."],
            ["Saya sedang membangun demo embedding sederhana."],
        ],
        inputs=[text_in],
        label="Contoh",
    )

    btn.click(embed, inputs=[text_in, normalize], outputs=[vec_out, dim_out])

# Enable queue for concurrency
demo.queue()

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