PyTorch Vision Transformer (ViT) for Image Classification

This releases PyTorch W8A8 optimized weight and quantization config file for vision transformer model (ViT) for image classification tasks using ImageNet dataset