app.py

import gradio as gr
# from transformers import AutoBackbone, AutoModelForImageClassification, AutoImageProcessor, Swinv2ForImageClassification
from transformers import pipeline, AutoImageProcessor, Swinv2ForImageClassification, Swinv2Model
from torchvision import transforms

# model = AutoModelForImageClassification.from_pretrained("haywoodsloan/ai-image-detector-deploy")
# image_processor = AutoImageProcessor.from_pretrained("haywoodsloan/ai-image-detector-deploy")
image_processor = AutoImageProcessor.from_pretrained("haywoodsloan/ai-image-detector-deploy")
# image_processor = Swinv2ForImageClassification.from_pretrained("haywoodsloan/ai-image-detector-deploy")
model = Swinv2ForImageClassification.from_pretrained("haywoodsloan/ai-image-detector-deploy")

clf = pipeline(model=model, task="image-classification", image_processor=image_processor)

class_names = ['artificial', 'real']

def predict_image(img):
  img = transforms.ToPILImage()(img)
  img = transforms.Resize((256,256))(img)
  prediction=clf.predict(img)
  return {class_names[i]: float(prediction[i]["score"]) for i in range(2)}

image = gr.Image(label="Image to Analyze", sources=['upload'])
label = gr.Label(num_top_classes=2)

gr.Interface(fn=predict_image, inputs=image, outputs=label, title="AI Generated Classification").launch()