Fix #14 app.py

app.py

@@ -13,31 +13,24 @@ from huggingface_hub import hf_hub_download
 class ResBlk(nn.Module):
     def __init__(self, dim_in, dim_out, normalize=False, downsample=False):
         super().__init__()
-        self.conv1 = nn.Conv2d(dim_in, dim_out, 3, 1, 1)
-        self.norm1 = nn.InstanceNorm2d(dim_out, affine=True) if normalize else None
-        self.relu1 = nn.ReLU(inplace=True)
-        self.conv2 = nn.Conv2d(dim_out, dim_out, 3, 1, 1)
-        self.norm2 = nn.InstanceNorm2d(dim_out, affine=True) if normalize else None
-        self.relu2 = nn.ReLU(inplace=True)
+        self.normalize = normalize
         self.downsample = downsample
-        if self.downsample:
-            self.avg_pool = nn.AvgPool2d(2)
+        self.main = nn.Sequential(
+            nn.Conv2d(dim_in, dim_out, 3, 1, 1),
+            nn.InstanceNorm2d(dim_out, affine=True) if normalize else nn.Identity(),
+            nn.ReLU(inplace=True),
+            nn.Conv2d(dim_out, dim_out, 3, 1, 1),
+            nn.InstanceNorm2d(dim_out, affine=True) if normalize else nn.Identity()
+        )
+        self.downsample_layer = nn.AvgPool2d(2) if downsample else nn.Identity()
+        self.skip = nn.Conv2d(dim_in, dim_out, 1, 1, 0, bias=False)
 
     def forward(self, x):
-        residual = x
-        out = self.conv1(x)
-        if self.norm1:
-            out = self.norm1(out)
-        out = self.relu1(out)
-        out = self.conv2(out)
-        if self.norm2:
-            out = self.norm2(out)
-        out = self.relu2(out)
-        if self.downsample:
-            out = self.avg_pool(out)
-            residual = self.avg_pool(residual)
-        out = out + residual
-        return out
+        out = self.main(x)
+        out = self.downsample_layer(out)
+        skip = self.skip(x)
+        skip = self.downsample_layer(skip)
+        return (out + skip) / math.sqrt(2)
 
 class AdainResBlk(nn.Module):
     def __init__(self, dim_in, dim_out, style_dim=64, w_hpf=1, upsample=False):
@@ -109,8 +102,8 @@ class MappingNetwork(nn.Module):
         for layer in self.unshared:
             out += [layer(h)]
         out = torch.stack(out, dim=1) # (batch, num_domains, style_dim)
-        idx = torch.arange(y.size(0)).to(y.device)
-        s = out[idx, y]
+        idx = torch.LongTensor(range(y.size(0))).unsqueeze(1).to(y.device)
+        s = torch.gather(out, 1, idx.unsqueeze(2).expand(-1, -1, out.size(2))).squeeze(1)
         return s
 
 class StyleEncoder(nn.Module):
@@ -177,8 +170,8 @@ class Generator(nn.Module):
 
 # FUNCIÓN PARA CARGAR EL MODELO
 def load_pretrained_model(ckpt_path, img_size=256, style_dim=64, num_domains=3, device='cpu'):
-    num_domains_mappin = 3
-    latent_dim_for_mapping = 13
+    num_domains_mappin = 2
+    latent_dim_for_mapping = 14
     G = Generator(img_size, style_dim).to(device)
     M = MappingNetwork(latent_dim_for_mapping, style_dim, num_domains).to(device)
     S = StyleEncoder(img_size, style_dim, num_domains).to(device)
@@ -235,7 +228,7 @@ if __name__ == '__main__':
     checkpoint_path = 'iter/12500_nets_ema.ckpt'
     img_size = 128
     style_dim = 64
-    num_domains = 3
+    num_domains = 2
     device = 'cuda' if torch.cuda.is_available() else 'cpu'
 
     try: