Vit Tiny Patch8 112.cosface Ms1mv3

🚀 Vision Transformer (ViT) for Face Recognition

A Vision Transformer (ViT) model designed for face recognition, trained on the MS1MV3 dataset. It offers seamless compatibility with the timm library.

🚀 Quick Start

This Vision Transformer (ViT) is specifically developed for face recognition and trained on the MS1MV3 dataset. The model was trained using the following repository: https://github.com/gau-nernst/timm-face. It is fully compatible with the timm library.

💻 Usage Examples

Basic Usage

import timm
import torch.nn.functional as F

model = timm.create_model("hf_hub:gaunernst/vit_tiny_patch8_112.cosface_ms1mv3", pretrained=True).eval()
embs = model(torch.randn(1, 3, 112, 112))  # output shape (1, 512)
embs = F.normalize(embs, dim=1)  # model output is not normalized

📚 Documentation

Model Information