Zhengyang Feng

Awaiting implementation details from the authors.

@solidexu I don't have spare gpu right now. I will try test it tomorrow.

> I don't know why, the issue is solved by commenting out the relu in RESAReducer. It's too STRANGE for me.  What pytorch version are you using & do...

@solidexu I don't really have 1.10, but I can start training normally with 1.6.0 (I have only one card so I first changed world_size to 1 and then use only...

Are you running customized code or do you see that error in the current master branch?

@Cverlpeng 由于第一季度进展较慢，PRNet已经被推迟到了第二季度计划（4.1 - 6.31）#4，目前有标注为 in progress都是计划中但尚未实现的方法。 --- Due to slow progress in Q1，PRNet has been postponed to Q2 (4.1 - 6.31) #4，all methods marked as **in progress** are scheduled but...

@Cverlpeng PRNet有很多关键的实现细节论文中没有透露，作者也一直不回复邮件，可能短期内是复现不出来了。 目前已经写的部分在这里： https://github.com/voldemortX/pytorch-auto-drive/tree/prnet

Your apex installation could be incorrect. Although it is only a warning, I am not sure if this won't affect mix precision training.

@wing212 Hi! Did you run code from the latest master branch? And could I ask what is your exact setup? e.g. PyTorch & TorchVision version, CUDA version, whether mixed precision...

Did you accidentally closed the issue?