makaichi

Fix Nan loss

2

I got "Nan" when use Mix loss to train (not speech denoise task), and Fix it by adding grad clip as fellows: loss.backward() nn.utils.clip_grad_norm_(self.estimator.parameters(), 10.0) # add this to clip...

Fix an error

1

I got an error (torch 1.10.0), and fix it by phase_conv2(Conv1d) ------> phase_conv2(Conv2d)

GRU shoule be set as batch_first=True?

1

The input shape to "PHASEN..rnn" is [B,T,D*C] as in PHASEN.forward, so the batch_first=True (by default batch_first=False) should be set?