Can you supply some training logs

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

1.46k stars 84 forks source link

Can you supply some training logs #89

Open powermano opened 1 month ago

powermano commented 1 month ago

Can you supply some training logs. I want to know whether my trianing process is correct