luost26 / diffusion-point-cloud

:thought_balloon: Diffusion Probabilistic Models for 3D Point Cloud Generation (CVPR 2021)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Question about training

yufeng9819 opened this issue · comments

Hi! I am interested in your wonderful work!

I have a question about training. In your training file "train_gen.py", you set the training iteration number to 'inf' and the training loss is around 60.5, also the grad is around 0.2 to 0.3. So I want to know when to stop straining and I cannot judge stop time just by the loss and grad. Also I want to know why the training loss almost ever decrease a lot.

Looking forward to your response!

Best.

Hi,

Thank you for your interest in our project. Could you take a look at the generated point clouds in the tensorboard? Sometimes the absolute value of loss does not tell anything.

I got it! Thanks for your response