app.py

import gradio as gr
from transformers import AutoProcessor, AutoModelForCausalLM
import spaces

import io
from PIL import Image
import base64  # Para decodificar imagens Base64
import subprocess
subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
#
model_id = 'J-LAB/Florence-vl3'
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True).to("cuda").eval()
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)

DESCRIPTION = "# Product Describe by Fluxi IA\n### Base Model [Florence-2] (https://huggingface.co/microsoft/Florence-2-large)"

@spaces.GPU
def run_example(task_prompt, image):
    inputs = processor(text=task_prompt, images=image, return_tensors="pt").to("cuda")
    generated_ids = model.generate(
        input_ids=inputs["input_ids"],
        pixel_values=inputs["pixel_values"],
        max_new_tokens=1024,
        early_stopping=False,
        do_sample=False,
        num_beams=3,
    )
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=False)[0]
    parsed_answer = processor.post_process_generation(
        generated_text,
        task=task_prompt,
        image_size=(image.width, image.height)
    )
    return parsed_answer

def process_image(image, task_prompt):
    if isinstance(image, str):
        if image.startswith('data:image/png;base64,'):
            # Decodifica a imagem Base64
            image_data = base64.b64decode(image.split(',')[1])
            image = Image.open(io.BytesIO(image_data))
    
    image = Image.fromarray(image)  # Convert NumPy array to PIL Image
    if task_prompt == 'Product Caption':
        task_prompt = '<MORE_DETAILED_CAPTION>'
    elif task_prompt == 'OCR':
        task_prompt = '<OCR>'
    
    results = run_example(task_prompt, image)
    
    # Remove the key and get the text value
    if results and task_prompt in results:
        output_text = results[task_prompt]
    else:
        output_text = ""

    # Convert newline characters to HTML line breaks
    output_text = output_text.replace("\n\n", "<br><br>").replace("\n", "<br>")

    return output_text

css = """
  #output {
    overflow: auto; 
    border: 1px solid #ccc; 
    padding: 10px;
    background-color: rgb(31 41 55);
    color: #fff; 
  }
"""

js = """
function adjustHeight() {
    var outputElement = document.getElementById('output');
    outputElement.style.height = 'auto';  // Reset height to auto to get the actual content height
    var height = outputElement.scrollHeight + 'px';  // Get the scrollHeight
    outputElement.style.height = height;  // Set the height
}

// Attach the adjustHeight function to the click event of the submit button
document.querySelector('button').addEventListener('click', function() {
    setTimeout(adjustHeight, 500);  // Adjust the height after a small delay to ensure content is loaded
});
"""

single_task_list =[
    'Product Caption', 'OCR'
]

with gr.Blocks(css=css) as demo:
    gr.Markdown(DESCRIPTION)
    with gr.Tab(label="Product Image Select"):
        with gr.Row():
            with gr.Column():
                input_img = gr.Image(label="Input Picture")
                task_prompt = gr.Dropdown(choices=single_task_list, label="Task Prompt", value="Product Caption")
                submit_btn = gr.Button(value="Submit")
            with gr.Column():
                output_text = gr.HTML(label="Output Text", elem_id="output")
        
        gr.Markdown("""
        ## How to use via API
        To use this model via API, you can follow the example code below:

        
python
        !pip install gradio_client
        from gradio_client import Client, handle_file

        client = Client("J-LAB/Fluxi-IA")
        result = client.predict(
            image=handle_file('https://raw.githubusercontent.com/gradio-app/gradio/main/test/test_files/bus.png'),
            api_name="/process_image"
        )
        print(result)

        """)
        
        submit_btn.click(process_image, [input_img, task_prompt], [output_text])

    demo.load(lambda: None, inputs=None, outputs=None, js=js)

demo.launch(debug=True)