After clicking ‘Stop Training’ you can simply run the ‘train’ function because ‘train’ updates the agent at each episode. So, you can resume training by calling ‘train(agent,env,trainOpts)’ again, without losing the trained parameters learned during the first call to train. You can simply begin the training.
If you want to save candidate agents before the fulfilment of your stop conditions even though training has not finished, the ‘rlTrainingOptions’ allows you to perform that. You can refer to the documentation below to understand how to use that function.
Thank you.
