jeremoo/

vit-fef-finetuned

For EmoryHacks:

I used an NVIDIA H100 through a DigitalOcean GPU Droplet to finetune Google’s Vision Transformer (ViT) model pre-trained on ImageNet-21k using a Facial Expression of Fatigues (FEF) dataset from Kaggle.

Base Model: https://huggingface.co/google/vit-base-patch16-224

Dataset: https://www.kaggle.com/datasets/ismailnasri20/driver-drowsiness-dataset-ddd