app.py

import gradio as gr
from gradio_client import Client

#fusecap_client = Client("https://noamrot-fusecap-image-captioning.hf.space/")
fuyu_client = Client("https://adept-fuyu-8b-demo.hf.space/")

def get_caption(image_in):
    
    fuyu_result = fuyu_client.predict(
	    image_in,	# str representing input in 'raw_image' Image component
	    True,	# bool  in 'Enable detailed captioning' Checkbox component
		fn_index=2
    )

    # Find the last occurrence of "."
    last_period_index = fuyu_result.rfind('.')

    # Truncate the string up to the last period
    truncated_caption = fuyu_result[:last_period_index + 1]

    # print(truncated_caption)
    print(f"\n—\nIMAGE CAPTION: {truncated_caption}")
    
    return truncated_caption
 

def infer(image_in):
    gr.Info("Getting image caption with Fuyu...")
    user_prompt = get_caption(image_in)
    
    
    
    return user_prompt

title = f"LLM Agent from a Picture",
description = f"Get a LLM system prompt from a picture so you can use it in <a href='https://huggingface.co/spaces/abidlabs/GPT-Baker'>GPT-Baker</a>."

css = """
#col-container{
    margin: 0 auto;
    max-width: 780px;
    text-align: left;
}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.HTML(f"""
        <h2 style="text-align: center;">Words from a Picture</h2>
        <p style="text-align: center;">{description}</p>
        """)
        
        with gr.Row():
            with gr.Column():
                image_in = gr.Image(
                    label = "Image reference",
                    type = "filepath",
                    elem_id = "image-in"
                )
                submit_btn = gr.Button("Make desciptions of my pic !")
            with gr.Column():
                result = gr.Textbox(
                    label ="Suggested System",
                    lines = 6,
                    max_lines = 30,
                    elem_id = "suggested-system-prompt"
                )
        with gr.Row():
            gr.Examples(
                examples = [
                    ["examples/ponder.png"],
                    ["examples/ponder2.png"],
                     
                ],
                fn = infer,
                inputs = [image_in],
                outputs = [result],
                cache_examples = True
            )

    submit_btn.click(
        fn = infer,
        inputs = [
            image_in
        ],
        outputs =[
            result
        ]
    )

demo.queue().launch(show_api=False)