I have been using deeplabcut for about 3 weeks, and after asking around on forums, I was told that DLC GUI does not have a way to continue training the network from a saved snapshot. That is, I have a snapshot of a network at 650 epochs, and I would like to train it for an additional 100 epochs, to 750. There is a way to do this in the terminal, but I use the GUI, and it seems like it's an important feature to have in both interfaces. There used to be a way to do it in the pose.cfg file with tensorflow, but I tried something similar with the new pytorch version to no avail. I think there should be an option to load a snapshot file path and continue training after that snapshot so that users can improve upon network training rather than having to start from 0 epochs each time more training is desired.