Jesse Yang

[This](https://github.com/CharlesShang/DCNv2/pull/50) adds apex support with level O1. But I got the following error when running it. `RuntimeError: Function _DCNv2Backward returned an invalid gradient at index 1 - expected type torch.cuda.HalfTensor...

The problem is solved. Besides following the code of [https://github.com/lbin/DCNv2](url), I add the following three lines before `return` in the _backward function of _DCNv2 in dcn_v2.py: ``` grad_input = grad_input.half()...

I followed the code in this branch and got the following error: `RuntimeError: Function _DCNv2Backward returned an invalid gradient at index 1 - expected type torch.cuda.HalfTensor but got torch.cuda.FloatTensor`

Thanks for your reply. I don't think recompiling can solve this problem. I re-clone the code from [https://github.com/lbin/DCNv2](url) and build it, but still got the same error. My pytorch version...

The problem is solved. I add the following three lines before `return` in the _backward function of _DCNv2 in dcn_v2.py: ``` grad_input = grad_input.half() grad_offset = grad_offset.half() grad_mask = grad_mask.half()...

I have implemented YOLOv2 in tensorflow. But I can achieve an mAP of about only 0.60 on VOC07-test (train with VOC07+12 train+val), with all the tricks except "hi-res detector" in...

@cory8249 I see. Thanks!

@cory8249 Have you fixed [another issue](https://github.com/longcw/yolo2-pytorch/issues/18) when you got the 0.6825 mAP?

@cory8249 In my understanding, the l.delta in darknet source code is the minus derivative of the loss with respect to the input value. If the mask for those positions without...