after shuffle buffer filled, the training seems stopped

chrisdonahue / wavegan

WaveGAN: Learn to synthesize raw audio with generative adversarial networks

MIT License

1.33k stars 281 forks source link

after shuffle buffer filled, the training seems stopped #26

Closed klauspa closed 5 years ago

klauspa commented 5 years ago

I train the wavegan model using a single 1080Ti Nvidia Graphic card, 12 GB memory, After the screen shows shuffle buffer filled, the training seems stopped. It just stuck there for more than forty hours.

pankajbadatia commented 5 years ago

Its not stopped. check your gpu utilization.
(1) tensorboard --logdir=./train
use this for looking the training steps in tensorboard. (2) Incase the length of wav file is not (16384, 32768, or 65536) use --data_first_slice and --data_pad_end as a true flag while training .

pankajbadatia commented 5 years ago

close this issue incase your problems has solved.