app.py

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

import subprocess
subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)

models = {
    'gokaygokay/Florence-2-Flux-Large': AutoModelForCausalLM.from_pretrained('gokaygokay/Florence-2-Flux-Large', trust_remote_code=True).eval(),
    'gokaygokay/Florence-2-Flux': AutoModelForCausalLM.from_pretrained('gokaygokay/Florence-2-Flux', trust_remote_code=True).eval(),
}

processors = {
    'gokaygokay/Florence-2-Flux-Large': AutoProcessor.from_pretrained('gokaygokay/Florence-2-Flux-Large', trust_remote_code=True),
    'gokaygokay/Florence-2-Flux': AutoProcessor.from_pretrained('gokaygokay/Florence-2-Flux', trust_remote_code=True),
}

title = """<h1 align="center">Florence-2 Captioner for Flux Prompts</h1>
<p><center>
<a href="https://huggingface.co/gokaygokay/Florence-2-Flux-Large" target="_blank">[Florence-2 Flux Large]</a>
<a href="https://huggingface.co/gokaygokay/Florence-2-Flux" target="_blank">[Florence-2 Flux Base]</a>
</center></p>
"""

@spaces.GPU
def run_example(image, model_name='gokaygokay/Florence-2-Flux-Large'):
    image = Image.fromarray(image)
    task_prompt = "<DESCRIPTION>"
    prompt = task_prompt + "Describe this image in great detail."

    if image.mode != "RGB":
        image = image.convert("RGB")

    model = models[model_name]
    processor = processors[model_name]

    inputs = processor(text=prompt, images=image, return_tensors="pt")
    generated_ids = model.generate(
        input_ids=inputs["input_ids"],
        pixel_values=inputs["pixel_values"],
        max_new_tokens=1024,
        num_beams=3,
        repetition_penalty=1.10,
    )
    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["<DESCRIPTION>"]

with gr.Blocks(theme='bethecloud/storj_theme') as demo:
    gr.HTML(title)

    with gr.Row():
        with gr.Column():
            input_img = gr.Image(label="Input Picture")
            model_selector = gr.Dropdown(choices=list(models.keys()), label="Model", value='gokaygokay/Florence-2-Flux-Large')
            submit_btn = gr.Button(value="Submit")
        with gr.Column():
            output_text = gr.Textbox(label="Output Text")

    gr.Examples(
        [["image1.jpg"], 
         ["image2.jpg"], 
         ["image3.png"], 
         ["image5.jpg"]],
        inputs=[input_img, model_selector],
        outputs=[output_text],
        fn=run_example,
        label='Try captioning on below examples',
        cache_examples=True
    )

    submit_btn.click(run_example, [input_img, model_selector], [output_text])

demo.launch(debug=True)