model_utils.py

import torch
from transformers import ViTForImageClassification, ViTImageProcessor

def load_model_and_processor(model_path, model_name_or_path, class_names):
    """
    Loads the ViT model and processor.
    """
    processor = ViTImageProcessor.from_pretrained(model_name_or_path)
    
    model = ViTForImageClassification.from_pretrained(
        model_path,
        num_labels=len(class_names),
        id2label={str(i): label for i, label in enumerate(class_names)},
        label2id={label: i for i, label in enumerate(class_names)},
    )
    model.eval()
    return model, processor

def predict(model, processor, img, device="cpu"):
    """
    Runs inference on an image and returns logits, probabilities, and prediction.
    """
    img = img.convert("RGB")
    processed_input = processor(images=img, return_tensors="pt").to(device)
    pixel_values = processed_input["pixel_values"].to(device)

    with torch.no_grad():
        outputs = model(pixel_values, output_attentions=True)
        logits = outputs.logits
        probabilities = torch.softmax(logits, dim=1)[0].tolist()
        prediction = torch.argmax(logits, dim=-1).item()
    
    return outputs, processed_input, probabilities, prediction