app.py

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


device='cpu'

processor = AutoProcessor.from_pretrained("microsoft/git-base")
model = AutoModelForCausalLM.from_pretrained("nkasmanoff/git-planet").to(device)



def predict(image,max_length=64,device='cpu'):
    pixel_values = processor(images=image, return_tensors="pt").to(device).pixel_values
    generated_ids = model.generate(pixel_values=pixel_values, max_length=max_length)
    generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
    
    return generated_caption 



input = gr.inputs.Image(label="Upload your Image", type = 'pil', optional=True)
output = gr.outputs.Textbox(type="text",label="Captions")


title = "Image Captioning"

interface = gr.Interface(
        fn=predict,
        inputs = input,
        theme="grass",
        outputs=output,
        title=title,
    )
interface.launch(debug=True)