ICCV2021, Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
Primary LanguageJupyter NotebookOtherNOASSERTION