Hi there,
First of all, thank you for sharing this great work. I am very interested in your SDT method and would like to reproduce the results reported in the paper (or test it on my own data).
However, I noticed that the SDT pre-trained weights are not currently available in the repository. Could you please share the pre-trained checkpoints?
Having the pre-trained models would be extremely helpful for the community to better understand and build upon your work.
Thanks again for your contribution!