search

OpenGVLab / Vision-RWKV

Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architectures
https://arxiv.org/abs/2403.02308
Apache License 2.0
371 stars 14 forks source link

MAE pretraining #22

Closed sahilqure closed 2 weeks ago

sahilqure commented 5 months ago

It will be really nice if you could provide us the code for MAE pretraining.

duanduanduanyuchen commented 4 months ago

It will be really nice if you could provide us the code for MAE pretraining.

Hi, we follow the code of facebookresearch/mae for MAE pertaining. The token shift is changed to a 1D direction in the model.

duanduanduanyuchen commented 2 weeks ago

Hi, the mae code is uploaded. Please refer to this link: mae code.