---
library_name: keras
---

This model is a TensorFlow port of ViT B-16 [1] trained with recipes from [2]. It was first pre-trained on ImageNet-21k and was then fine-tuned on the ImageNet-1k dataset. You can refer to [this notebook](https://github.com/sayakpaul/probing-vits/blob/main/notebooks/load-jax-weights-vitb16.ipynb) to know how the porting was done.

## References

[1] An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: https://arxiv.org/abs/2010.11929

[2] How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers: https://arxiv.org/abs/2106.10270