Continuing train that stopped

[Yd74]

Does it possible to continue scaling-and-training if the training stopped in the middle of work? Or I have to remove the training folder and start from the beginning?

[singulart]

This is not possible. I had to fork the repo and implement continuations myself: https://github.com/singulart/SinGAN

[yeeunsong]

This is not possible. I had to fork the repo and implement continuations myself: https://github.com/singulart/SinGAN

Hello, can I ask a question about continuing train that has been stopped? Could you please check if I understood what you had said?

SinGAN currently in this repo does not support continuations. But repo here (https://github.com/singulart/SinGAN) implements continuations.

Thank you!

[yeeunsong]

This is not possible. I had to fork the repo and implement continuations myself: https://github.com/singulart/SinGAN

So if I want to continue the training that has stopped, can I just execute the command like the below? (The same as the start command of the training)

python main_train.py --input_name <input_file_name>

[ageroul]

This is not possible. I had to fork the repo and implement continuations myself: https://github.com/singulart/SinGAN

Thanks @tamarott for the great work! @singulart thanks also for the "hack", it works great, especially useful when working with Colab. I had one issue with your implementation when I tried to execute the "main.train" with another (a second) picture. I realised I had to delete the opt.db file first because it remembers the previous state (from the the first) picture. So every time I need to run the main_train.py for a new picture, I first delete the opt.db file.

Thanks,