The finetuning weights on imagenet

keyu-tian / SparK

[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"

https://arxiv.org/abs/2301.03580

MIT License

1.42k stars 82 forks source link

The finetuning weights on imagenet #28

Open YuetongFang opened 1 year ago

YuetongFang commented 1 year ago

Thanks for your wonderful work! The tutorial is fairly friendly for reproduction, but I also wonder if it is possible for us to get the fine-tuned weights on imagenet?

keyu-tian commented 1 year ago

yes, i think those would be ok in a couple of weeks. i'll work on them (weights & logs like in https://github.com/keyu-tian/SparK/tree/main/downstream_d2#fine-tuned-resnet-50-weights-log-files-and-performance) once i finish my current project at hand.