Update src/sbv2/generator.py

src/sbv2/generator.py

@@ -1,33 +1,20 @@
-import torch
-from torch import nn
-import torch.nn.functional as F
-from .modules import LayerNorm, ConvReluNorm
-
-class Generator(nn.Module):
-    def __init__(self, channels):
-        super(Generator, self).__init__()
-        self.conv_pre = nn.Conv1d(channels, 512, 7, 1, 3)
-        self.resblocks = nn.ModuleList([
-            ResBlock(512) for _ in range(3)
-        ])
-        self.conv_post = nn.Conv1d(512, 1, 7, 1, 3)
-
-    def forward(self, x):
-        x = self.conv_pre(x)
-        for resblock in self.resblocks:
-            x = resblock(x)
-        x = self.conv_post(x)
-        x = torch.tanh(x)
-        return x
-
-class ResBlock(nn.Module):
-    def __init__(self, channels):
-        super(ResBlock, self).__init__()
-        self.convs = nn.Sequential(
-            nn.Conv1d(channels, channels, 3, 1, 1),
-            nn.ReLU(),
-            nn.Conv1d(channels, channels, 3, 1, 1)
-        )
-
-    def forward(self, x):
-        return x + self.convs(x)
+import torch.nn as nn
+
+class Generator(nn.Module):
+    def __init__(
+        self,
+        upsample_rates,
+        upsample_initial_channel,
+        resblock_kernel_sizes,
+        resblock_dilation_sizes,
+        resblock,
+        upsample_kernel_sizes,
+        inter_channels,
+        out_channels,
+        sampling_rate
+    ):
+        super().__init__()
+        self.dummy = nn.Identity()  # 実装は後で拡張可
+
+    def forward(self, x):
+        return self.dummy(x)