Closed iamanigeeit closed 2 years ago
Good catch. Would you send a PR including these changes?
@erogol Unfortunately i cloned the repo in Oct and made a bunch of changes along the way... here's the new config file: parallel_wavegan_config.txt
Sorry i just saw the comments on my PR!
The configs in the original paper are as mentioned earlier
stft_loss_weight=1.0, # currently 0.5
mse_G_loss_weight=4.0, # currently 2.5
steps_to_start_discriminator=100000, # currently 200000
lr_gen=0.0001, # currently 0.0002
lr_disc=0.00005, # currently 0.0002
Original paper section 4.1.2:
The hyper-parameter λ adv in equation (7) was chosen to be 4.0 based on our preliminary experiments.
Note that the discriminator was fixed for the first 100K steps, and two models were jointly trained afterwards.
The initial learning rate was set to 0.0001 and 0.00005 for the generator and discriminator, respectively.
If the current config works better than i'm ok with it.
Hi,
I have tried training with the current default config and it drops the learning rate too fast and the model converges to generate noise. Because when
scheduler_after_epoch=False
,ExponentialLR
withgamma=0.999
will cause the learning rate to reach 0.000 in 10k steps.Config from the original paper are:
It is also possible to use ExponentialLR with some float rounding error:
With more GPU memory the batch_size can be increased and steps reduced.