⚗️ | Experiment with other layer layouts.

generator.py

@@ -3,21 +3,25 @@ import torch.nn as nn
 class SISUGenerator(nn.Module):
     def __init__(self, upscale_scale=1): # No noise_dim parameter
         super(SISUGenerator, self).__init__()
-        self.model = nn.Sequential(
+        self.layers1 = nn.Sequential(
             nn.Conv1d(2, 128, kernel_size=3, padding=1),
-            nn.LeakyReLU(0.2, inplace=True),
+            # nn.LeakyReLU(0.2, inplace=True),
             nn.Conv1d(128, 256, kernel_size=3, padding=1),
-            nn.LeakyReLU(0.2, inplace=True),
-
-            nn.Upsample(scale_factor=upscale_scale, mode='nearest'),
-
-            nn.Conv1d(256, 128, kernel_size=3, padding=1),
-            nn.LeakyReLU(0.2, inplace=True),
-            nn.Conv1d(128, 64, kernel_size=3, padding=1),
-            nn.LeakyReLU(0.2, inplace=True),
-            nn.Conv1d(64, 2, kernel_size=3, padding=1),
-            nn.Tanh()
+            # nn.LeakyReLU(0.2, inplace=True),
         )
 
-    def forward(self, x):
-        return self.model(x)
+        self.layers2 = nn.Sequential(
+            nn.Conv1d(256, 128, kernel_size=3, padding=1),
+            # nn.LeakyReLU(0.2, inplace=True),
+            nn.Conv1d(128, 64, kernel_size=3, padding=1),
+            # nn.LeakyReLU(0.2, inplace=True),
+            nn.Conv1d(64, 2, kernel_size=3, padding=1),
+            # nn.Tanh()
+        )
+
+    def forward(self, x, scale):
+        x = self.layers1(x)
+        upsample = nn.Upsample(scale_factor=scale, mode='nearest')
+        x = upsample(x)
+        x = self.layers2(x)
+        return x