update

toolbox/torchaudio/models/spectrum_dfnet/modeling_spectrum_dfnet.py

@@ -807,6 +807,7 @@ class SpectrumDfNet(nn.Module):
         # feat_power shape: [batch_size, spec_bins, time_steps]
         # feat_power shape: [batch_size, 1, spec_bins, time_steps]
         # feat_power shape: [batch_size, 1, time_steps, spec_bins]
+        feat_power = feat_power.detach()
 
         # spec shape: [batch_size, spec_bins, time_steps]
         feat_spec = torch.view_as_real(spec_complex)
@@ -815,14 +816,16 @@ class SpectrumDfNet(nn.Module):
         # feat_spec shape: [batch_size, 2, time_steps, spec_bins]
         feat_spec = feat_spec[..., :self.df_decoder.df_bins]
         # feat_spec shape: [batch_size, 2, time_steps, df_bins]
-
-        # spec shape: [batch_size, spec_bins, time_steps]
-        spec = torch.unsqueeze(spec_complex, dim=1)
-        # spec shape: [batch_size, 1, spec_bins, time_steps]
-        spec = spec.permute(0, 1, 3, 2)
-        # spec shape: [batch_size, 1, time_steps, spec_bins]
-        spec = torch.view_as_real(spec)
-        # spec shape: [batch_size, 1, time_steps, spec_bins, 2]
+        feat_spec = feat_spec.detach()
+
+        # # spec shape: [batch_size, spec_bins, time_steps]
+        # spec = torch.unsqueeze(spec_complex, dim=1)
+        # # spec shape: [batch_size, 1, spec_bins, time_steps]
+        # spec = spec.permute(0, 1, 3, 2)
+        # # spec shape: [batch_size, 1, time_steps, spec_bins]
+        # spec = torch.view_as_real(spec)
+        # # spec shape: [batch_size, 1, time_steps, spec_bins, 2]
+        # spec = spec.detach()
 
         e0, e1, e2, e3, emb, c0, lsnr, h = self.encoder.forward(feat_power, feat_spec)