Jinhong Deng

@yujheli Thanks for your reply. Yes, I see the large improvements in the setting of clipart and watercolor using Resnet-101 as the backbone. Due to previous works using vgg16 without...

@yujheli There is a small bug in: https://github.com/facebookresearch/adaptive_teacher/blob/cba3c59cadfc9f1a3a676a82bf63d76579ab552b/adapteacher/modeling/meta_arch/rcnn.py#L246 This branch `supervised_target ` is never executed although it does not influence the results because the adversarial loss weight is quite small.

@yujheli Figure 2 in the paper shows that student networks adopt strong augmentation data to utilize adversarial training. However, in the code: https://github.com/facebookresearch/adaptive_teacher/blob/cba3c59cadfc9f1a3a676a82bf63d76579ab552b/adapteacher/engine/trainer.py#L612 The adversarial loss is calculated according to...

@tmp12316 I think there are serval reasons for high source-only results: 1. As you mentioned, the model with BN may benefit from the large batch size. 2. Detectron implementation usually...

@yujheli Thanks for your reply. I also try to reproduce the results where the vgg16 does not contain BN layers, but I encounter an error when the code starts the...

@yujheli Thanks for your reply. I am looking forward to seeing your update.

@yujheli Hi, I find the number of target samples is more than the one of the source domain on cityscapes to foggy cityscapes.  I guess you use all the...

At first, you can `fork ` this project to your own repository and then you can add your ideas to this project at your repository. you can pull an request...

I add this, but a new error occurs: ``` NameError: name 'numpy_include' is not defined ``` Thus how to define the `numpy_include`?

@123zhen123 The most serious foggy level, i.e. 0.02