Continue training from a checkpoint

simontomaskarlsson / GAN-MRI

Code repository for Frontiers article 'Generative Adversarial Networks for Image-to-Image Translation on Multi-Contrast MR Images - A Comparison of CycleGAN and UNIT'

GNU General Public License v3.0

227 stars 77 forks source link

Continue training from a checkpoint #17

Open Erfandarzi opened 4 years ago

Erfandarzi commented 4 years ago

Hey @simontomaskarlsson , Thanks for your great job! Could you please guide me on how to continue training from an epoch, using saved .hdf5 models? Thanks!

simontomaskarlsson commented 4 years ago

Hi @Erfandarzi, I would suggest using the keras function model.load_weights("path_to_weights") before line 239, and maybe force the correct epoch number to start on