ThibaultGROUEIX

I hit the same issue and had to downgrade opencv. The `[blender_path]/pip3 install -r requirements.txt` installs opencv 4.4 which is not retrocompatible for some reason. It triggers an issue with...

Can you share your minimal exemple @zhangmozhe ? I am unable to reproduce. As you can see from the example below, the output of geomLoss seems to differ from @colin97...

Thanks @zhangmozhe Your code points out that I forgot `torch.sqrt` in mine in Colin97's emd, so i am updated my above snippet just in case. However I am not sure...

Thanks @zhangmozhe !

Thanks ! Sorry i didn't dare dive in the details of the ELBO loss, i am not familiar with it and I would need comments to follow through. I just...

Related question: I am trying the code on AFFNIST and it systematically crashes. It seems that `z_sigma_2log` becomes too large [here](https://github.com/GuHongyang/VaDE-pytorch/blob/master/model.py#L172 ) which gives NaN after the exponential. Do you...

No I did not solve it. As you can see from table 2 [here](https://arxiv.org/pdf/2006.11132.pdf), we reported the divergence on problem on Affnist, but the code ran on our other toy...

Hi pxing6, Thanks for your interest! Have you tried all these solutions ? https://github.com/ThibaultGROUEIX/AtlasNet#a-note-on-data Cheers -Thibault

Hi, is this from a previous version of the repo? Best -Thibault

HI, I don't support this version anymore. You can try these links renderings : https://cloud.enpc.fr/s/axiC3FHS374FgEf points : https://cloud.enpc.fr/s/JNf3NAxGbQoQsKY Cheers, Thibault