import gradio as gr

examples = [
    ["forest", "https://ak.picdn.net/shutterstock/videos/21179416/preview/stock-footage-aerial-shot-winter-forest.mp4"],
    ["people", "https://ak.picdn.net/shutterstock/videos/5629184/preview/stock-footage-senior-couple-looking-through-binoculars-on-sailboat-together-shot-on-red-epic-for-high-quality-k.mp4"],
    ["orange", "https://ak.picdn.net/shutterstock/videos/1063125190/preview/stock-footage-a-beautiful-cookie-with-oranges-lies-on-a-green-tablecloth.mp4"]
]

def predict(labels, video):
    return labels, labels

with gr.Blocks() as demo:
    gr.Markdown(
        """
        # Ego-VPA Demo
        Choose a sample video and click predict to view the results.
        """
    )
    with gr.Row():
        with gr.Column():
            video = gr.PlayableVideo(label="video", interactive=False)
        with gr.Column():
            labels = gr.Text(label="Ground Truth")
            zeroshot = gr.Text(label="LaViLa (zero-shot) prediction")
            ours = gr.Text(label="Ego-VPA prediction")
    btn = gr.Button("Predict", variant="primary")
    btn.click(predict, inputs=[labels, video], outputs=[zeroshot, ours])
    gr.Examples(examples=examples, inputs=[labels, video])

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