980202006
980202006
I want to read multiple windows from a numpy file, like dividing a tensor into n equal parts.
it is because of the function timedistributed previous_mask the methods of resolution can be refer to https://blog.csdn.net/songbinxu/article/details/80242211
But the program can not do checkpointing, and Tensorboard summary logging, because of the steps.
> > That is great. > > How did you calculate F0 reconstruction loss? > > @kan-bayashi I used this: https://github.com/maxrmorrison/torchcrepe > Downsample the synthesized wave to 16k Hz (the...
@bigmindapp I had the same problem and was about to retrain, did you solve the problem?
@patil-suraj Thank you! Do you know the approximate loss of vae convergence? The vae loss I trained is very large.
I would be appreciate if you share it me too. My mail is [email protected]
大佬们都微调了多久,我微调后全是重复的语句