TsingW

@elvindp > 3. Without the score, if the target is out of FOV (i.e. out of image), the lightning model will always product a wrong output coordinate. Yeah, it is...

lightning的效果跟本没法落地

> > lightning的效果跟本没法落地 > > Really? There are many ways to optimize a research model for practical deployment. Looking forward to seeing your advance. 因为Lightning模型设计上并没有输出目标的置信度，哪怕是像ST那样有一个score模块都能勉强用。对于实际使用来说，Lightning这样的设计在找不到目标时只会胡乱给一个框，从模型训练的角度来看，它必然会给出一个尽可能大的框来获得更小的损失，而这种设计对落地来说毫无意义。当然在有限的场景下，Lightning优化速度的设计还是比较不错的。

> @ayanasser Hi, STARK can predict a confidence score besides the bounding box. You can set a threshold such as 0.5 to control the existence of the box. For example,...

Thanks! Here come another confusing lines, especially `Line#392`: https://github.com/IDEA-opensource/DAB-DETR/blob/309f6ad92af7a62d7732c1bdf1e0c7a69a7bdaef/models/dab_deformable_detr/deformable_transformer.py#L390-L392 Why concat `src_valid_ratios` with itself, and then multiplied to `reference_points`? When I alter the param `num_feature_levels` to another one, like 3,...

> Use torch.fft.irfft() instead of torch.rfft() But that is just the beginning of your problems in trying to run pytracking Agree. So much grateful to pytracking but the code of...

? 你这个问题有点哲学

@whh8162880 这个你能做的话欢迎提pr哦

需要的可以到 @jslyxyc 的分支上看看

可能是uiautomator框架和emui兼容性不好吧