Roland

I'm wondering about this as well. 1.414 is approximately sqrt(2), which is the diagonal of a unit square. Perhaps they are normalizing by the diagonal here? But their doc (https://github.com/NVlabs/eg3d/blob/main/docs/training_guide.md...

I have the same problem. I think it fails when the input size is too big

I see your point. It would be bad to lose a checkpoint one wants. I train my models on a slurm server that often preempts and requeues my jobs, so...

Thanks! I will try writing a callback myself.