Continue training from a checkpoint

[Erfandarzi]

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]

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