app.py

import torch
from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import gradio as gr
import io

# Load model and feature extractor
def load_model():
    processor = AutoImageProcessor.from_pretrained("therealcyberlord/stanford-car-vit-patch16")
    model = AutoModelForImageClassification.from_pretrained("therealcyberlord/stanford-car-vit-patch16")
    return processor, model

processor, model = load_model()

# Function to classify image
def classify_image(image):
    # Convert image if necessary
    if not isinstance(image, Image.Image):
        image = Image.open(io.BytesIO(image)).convert("RGB")
        
    inputs = processor(images=image, return_tensors="pt")
    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits
        predicted_class_idx = logits.argmax(-1).item()
        labels = model.config.id2label
        predicted_class = labels[predicted_class_idx]
    return predicted_class

# Define Gradio Interface
app = gr.Interface(
    fn=classify_image,
    inputs=gr.Image(type="pil"),
    outputs="text",
    title="Car Classification",
    description="Upload a car image to classify its model."
)

# Launch the app
app.launch(share=True)