I typically run a custom training loop and never associate my optimizer with my model(s) via the compile function. I would like to save my optimizer so I can more easily resume training. What is the best way to save an optimizer in its own file, independent of any model.
Hi @chasep255
Welcome to the TensorFlow Forum!
Could you please let us know if this issue still persists? If so, Please provide us some more details along with the reproducible code to understand the issue. Thank you!