You can deploy this model and start earning money today!

nwrim/
ViTNat

Name: ViTNat
Availability: InStock

safetensors

vit

apache-2.0

Views

No views yet

ViTNat

ViTNat is a fine-tuned Vision Transformer (ViT) model that predicts the naturalness of a scene as a continuous regression output. It is based on google/vit-base-patch16-224-in21k and was trained using naturalness ratings from human participants.

🔗 See more details in the GitHub repo: https://github.com/yourusername/ViTNat

📦 How to Use

python

from transformers import ViTImageProcessor, ViTForImageClassification
from PIL import Image

image = Image.open("your_image.jpg")

processor = ViTImageProcessor.from_pretrained("nwrim/ViTNat")
model = ViTForImageClassification.from_pretrained("nwrim/ViTNat", num_labels=1)

inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
print(outputs.logits)  # naturalness score

Note: Although the model class is ViTForImageClassification, it is configured for regression by setting num_labels=1. More info here.