Continue Training From Checkpoint

[Ebenezer319]

How is it advised to continue training a model from a checkpoint .pt file? Able to build out a simple process for predicting using a Pytorch based structure, but I am assuming there is a straight forward way to continue training like with Stable Baselines3. Correct me if I am wrong on this assumption.

Thanks in Advance!

[stellawang196]

Hi, the model will be loaded before and saved after the training process. You can change the args.cwd to the current working directory (where your models are saved) and then continue to train.