EsViT

[FrancescoSaverioZuppichini]

Paper

Implement Efficient Self-Supervised Vision Transformers (EsViT)

Link to the paper

TODOs

• [ ] develop the model in a scalable way (using *Block syntax and nn modules)
• [ ] draw the model on figma
• [ ] write doc
• [ ] test the model
• [ ] get (if possible) the pretrained weights