chrisdonahue / wavegan

WaveGAN: Learn to synthesize raw audio with generative adversarial networks
MIT License
1.33k stars 282 forks source link

Code for continue training model from last ckpt? #111

Open RitoRitoni opened 2 years ago

RitoRitoni commented 2 years ago

Can someone tell me how to continue training from last ckpt? Every time i use the folder where my last ckpt is stored i think it runs the model from scratch....

mattgallivan commented 2 years ago

It should start from where it's left off. Are you reading the shuffle buffer as it restarting? Because it must do that each time you train (resume or begin anew).