Terry Li

In my implementation, I refactored all of the NumPy operations to the PyTorch ones. It worked, and the accuracy can reach training from scratch.

Same problem, any solutions?

I tried the COCO dataset for testing, met the similar issue. The standard output as the following, ``` Traceback (most recent call last): File "test.py", line 141, in main() File...

Hi, all. A good news here. I tried Lite-HRNet model testing with code in this [repo](https://github.com/open-mmlab/mmpose) and it worked. Hope can help you.

The same issue from me. Is there any solutions?

Hi, hope this may help you. https://github.com/google/hdrnet/issues/4

The bilateral slicing operator is based on HDRNET and is not CPU-supported referenced from README file in HDRNET's repo.

is there any update on this issue?

Hi, Zruto. During my review, I was also puzzled by this problem. At some period of time, I checked a blog that mentioned this idea. The idea that use an...

Hi, @Zruto. I found it. Here is the blog. I hope can help you. https://blog.csdn.net/qq_36560894/article/details/108024748 Yep, literature [36] is referenced in the Zero-DCE paper. > Yuan and Sun [36] proposed...