Hi, so if I trained my model for 10 epochs and then saved its state_dict(). (And closed the laptop). So after loading the state_dict() and running the model again for 10 epochs. Will it be the same thing as running the model for 20 epochs at once?
Yes. If you use the optimizer containing estimates of gradients, such as Adam and AdamW, also try to save and load its
I am sorry, I don’t understand what you are saying.
This document might be helpful: Saving and Loading Models — PyTorch Tutorials 1.8.1+cu102 documentation.