removed docstrings from _init_weights

examples/baseline_models/unet.py

@@ -171,14 +171,6 @@ class DownsamplingBlock(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
-        
-        Parameters
-        -----------
-        m : torch.nn.Module
-            Module to initialize weights for
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=.02)
             if m.bias is not None:
@@ -344,14 +336,6 @@ class UpsamplingBlock(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
-        
-        Parameters
-        -----------
-        m : torch.nn.Module
-            Module to initialize weights for
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=.02)
             if m.bias is not None:

torch_harmonics/examples/models/s2segformer.py

@@ -117,14 +117,7 @@ class OverlapPatchMerging(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
 
-        Parameters
-        -----------
-        m : nn.Module
-            Module to initialize
-        """
         if isinstance(m, nn.LayerNorm):
             nn.init.constant_(m.bias, 0)
             nn.init.constant_(m.weight, 1.0)
@@ -230,14 +223,7 @@ class MixFFN(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
 
-        Parameters
-        -----------
-        m : nn.Module
-            Module to initialize
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=0.02)
             if m.bias is not None:
@@ -792,14 +778,7 @@ class SphericalSegformer(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
 
-        Parameters
-        -----------
-        m : nn.Module
-            Module to initialize
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=0.02)
             if m.bias is not None:

torch_harmonics/examples/models/s2unet.py

@@ -194,14 +194,7 @@ class DownsamplingBlock(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
 
-        Parameters
-        -----------
-        m : nn.Module
-            Module to initialize
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=0.02)
             if m.bias is not None:
@@ -585,14 +578,7 @@ class SphericalUNet(nn.Module):
         self.apply(self._init_weights)
 
     def _init_weights(self, m):
-        """
-        Initialize weights for the module.
 
-        Parameters
-        -----------
-        m : nn.Module
-            Module to initialize
-        """
         if isinstance(m, nn.Conv2d):
             nn.init.trunc_normal_(m.weight, std=0.02)
             if m.bias is not None: