app.py

from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
from PIL import Image
import gradio as gr

# Load the model, processor, and tokenizer
model = VisionEncoderDecoderModel.from_pretrained("microsoft/vision-encoder-decoder-base")
processor = ViTImageProcessor.from_pretrained("microsoft/vision-encoder-decoder-base")
tokenizer = AutoTokenizer.from_pretrained("microsoft/vision-encoder-decoder-base")

# Function to generate captions
def generate_caption(image):
    # Preprocess the image
    pixel_values = processor(images=image, return_tensors="pt").pixel_values
    
    # Generate caption
    output_ids = model.generate(pixel_values, max_length=16, num_beams=4)
    caption = tokenizer.decode(output_ids[0], skip_special_tokens=True)
    
    return caption

# Gradio interface
interface = gr.Interface(
    fn=generate_caption,
    inputs=gr.Image(type="pil"),
    outputs="text",
    title="Image to Text (Caption Generator)",
    description="Upload an image, and the AI will describe it!"
)

# Launch the interface
interface.launch()