app.py

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

# Load the pre-trained model, processor, and tokenizer
model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
feature_extractor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")

# Set the device to GPU if available
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

# Define generation parameters
max_length = 16
num_beams = 4

# Function to generate caption from image
def generate_caption(image):
    if image is None:
        return "Please upload an image."
    # Convert image to RGB if it's not
    if image.mode != "RGB":
        image = image.convert(mode="RGB")
    # Preprocess the image
    pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values
    pixel_values = pixel_values.to(device)
    # Generate caption
    output_ids = model.generate(pixel_values, max_length=max_length, num_beams=num_beams)
    caption = tokenizer.decode(output_ids[0], skip_special_tokens=True)
    return caption.strip()

# Create Gradio interface
iface = gr.Interface(
    fn=generate_caption,
    inputs=gr.Image(type="pil", label="Upload an Image"),
    outputs=gr.Textbox(label="Generated Caption"),
    title="🖼️ AI Image Caption Generator",
    description="Upload an image, and the AI will generate a descriptive caption for it.",
    allow_flagging="never"
)

# Launch the app
if __name__ == "__main__":
    iface.launch()