79824298

Date: 2025-11-19 10:02:13
Score: 1
Natty:
Report link

I found a better solution: I built a custom RNN where I store the sample vq losses in a Tensorarray and return this to the outer model, where the array is summed and the resulting loss is added to the model.

Reasons:
  • Whitelisted phrase (-2): solution:
  • Low length (0.5):
  • No code block (0.5):
  • Self-answer (0.5):
  • Single line (0.5):
  • Low reputation (1):
Posted by: Cola Lightyear