about checkpoint and performance for each category

[tpy001]

I am very interested in your great work. Your work shows that the Visual Foundation Model (VFM) has superior performance compared to other models in UDA tasks. However, i found the checkpoint is currently unavailable. Do you have any plans to upload the checkpoint file in the future?

In your paper, your model achieved an mIoU of 79.0% on GTAV->Cityscapes. But there is no IoU for each category, such as traffic light, car, person... I am curious about the IoU for each category.

Thanks a lot!