smiles724

I really agree with this proposal. ICLR becomes rather important in the AI field and should not be ignored.

Additionally, will you share both the dataset and training script so that we can reproduce your result? Thanks,

So, this is a reproduced code? Thanks anyway.

Hi, can I ask about the performance of your implemented Hengshuang method in part segmentation problem?

> I have write it "export PATH=/usr/local/cuda-11.0/bin:$PATH export LD_LIBRARY_PATH=/usr/local/cuda-11.0/lib64:$LD_LIBRARY_PATH" in ~/.bashrc, but it doesn't work, it's so weird. > > However, I can use geometric in terminal, but can't pycharm...

Agree with that point.