How to add more epochs and resume training (with optimizer state), after all training epochs are finished?

WongKinYiu / yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

GNU General Public License v3.0

13.41k stars 4.23k forks source link

How to add more epochs and resume training (with optimizer state), after all training epochs are finished? #2003

Open mattroos opened 8 months ago

mattroos commented 8 months ago

I'd like to continue training after all epochs have completed. The --resume switch will not do this even if I set --epochs to a higher number than during the previous training run. How can I do this? This is different that just starting a new training run and loading the weights from the last epoch of the previous training run because (as far as I know) that state of the optimizer is not utilized.