Can i release the reproduction code ?

[HannH]

Hi, I rewrite the DMFN model from your paper since i cant wait to see it's result. I want to know if i have the right to release the code and write a blog about your algorithm?

[Zheng222]

@HannH Of course, welcome to release your reproduction code and post.

[HannH]

@HannH Of course, welcome to release your reproduction code and post.

I have released the code. But I'm not sure if align loss is writed in the right way. Could you check it?
the code is here
https://github.com/HannH/DMFN/blob/2ade61431e243734a9de54c9770856a6fca9ba8c/model/loss.py#L77

[Zheng222]

@HannH
https://github.com/HannH/DMFN/blob/2ade61431e243734a9de54c9770856a6fca9ba8c/model/loss.py#L84-L87
c_u_k = sum_u_v(self.coord_x * tar / sum_tar)

The values of meshgrid should be [-1, 1].

[phongnhhn92]

@HannH have u checked this ? I am trying to re-implement this as well. I wonder if your code is correct or not.