Van Tiny

🚀 Van

The Van model is trained on ImageNet-1K. It offers a novel approach for image classification by leveraging a new attention layer.

🚀 Quick Start

The Van model is trained on ImageNet-1K. It was introduced in the paper Visual Attention Network and first released in this repository.

Disclaimer: The team releasing Van did not write a model card for this model so this model card has been written by the Hugging Face team.

✨ Features

• A new attention layer based on convolution operations is introduced, which can capture both local and distant relationships by combining normal and large kernel convolution layers. The large kernel convolution layer uses dilated convolution to capture distant correlations.

📚 Documentation

Model description

This paper introduces a new attention layer based on convolution operations able to capture both local and distant relationships. This is done by combining normal and large kernel convolution layers. The latter uses a dilated convolution to capture distant correlations.

Intended uses & limitations

You can use the raw model for image classification. See the model hub to look for fine - tuned versions on a task that interests you.

How to use

Here is how to use this model:

>>> from transformers import AutoFeatureExtractor, VanForImageClassification
>>> import torch
>>> from datasets import load_dataset

>>> dataset = load_dataset("huggingface/cats-image")
>>> image = dataset["test"]["image"][0]

>>> feature_extractor = AutoFeatureExtractor.from_pretrained("Visual-Attention-Network/van-base")
>>> model = VanForImageClassification.from_pretrained("Visual-Attention-Network/van-base")

>>> inputs = feature_extractor(image, return_tensors="pt")

>>> with torch.no_grad():
...     logits = model(**inputs).logits

>>> # model predicts one of the 1000 ImageNet classes
>>> predicted_label = logits.argmax(-1).item()
>>> print(model.config.id2label[predicted_label])
tabby, tabby cat

For more code examples, we refer to the documentation.

📄 License

This model is released under the Apache-2.0 license.