fixed bug - float network parameter rather than int/tuple

BrainMapperUNet.py

@@ -204,7 +204,7 @@ class BrainMapperUNet3D(nn.Module):
 
         self.bottleneck = modules.ConvolutionalBlock3D(parameters)
 
-        parameters['input_channels'] = parameters['output_channels'] * 2.0
+        parameters['input_channels'] = parameters['output_channels'] * 2
         self.decoderBlock1 = modules.DecoderBlock3D(parameters)
         self.decoderBlock2 = modules.DecoderBlock3D(parameters)
         self.decoderBlock3 = modules.DecoderBlock3D(parameters)
@@ -311,21 +311,23 @@ class BrainMapperUNet3D(nn.Module):
 
         return prediction
 
-# if __name__ == '__main__':
-
-#     # For debugging - To be deleted later! TODO
-
-#     parameters = {
-#         'kernel_heigth': 5,
-#         'kernel_width': 5,
-#         'kernel_classification': 1,
-#         'input_channels': 1,
-#         'output_channels': 64,
-#         'convolution_stride': 1,
-#         'dropout': 0.2,
-#         'pool_kernel_size': 2,
-#         'pool_stride': 2,
-#         'up_mode': 'upconv',
-#         'number_of_classes': 1
-#     }
-#     network = BrainMapperUNet(parameters)
+if __name__ == '__main__':
+
+    # For debugging - To be deleted later! TODO
+
+    parameters = {
+        'kernel_heigth': 5,
+        'kernel_width': 5,
+        'kernel_depth': 5,
+        'kernel_classification': 1,
+        'input_channels': 1,
+        'output_channels': 64,
+        'convolution_stride': 1,
+        'dropout': 0.2,
+        'pool_kernel_size': 2,
+        'pool_stride': 2,
+        'up_mode': 'upconv',
+        'number_of_classes': 1
+    }
+
+    network = BrainMapperUNet3D(parameters)

utils/modules.py

@@ -293,6 +293,7 @@ class ClassifierBlock(nn.Module):
 
         return logits
 
+
 class ConvolutionalBlock3D(nn.Module):
     """Parent class for a 3D convolutional block.
 
@@ -324,7 +325,7 @@ class ConvolutionalBlock3D(nn.Module):
         padding_depth = int((parameters['kernel_depth'] - 1) / 2)
 
         self.output_channels = parameters['output_channels']
-        self.convolutional_layer = nn.Conv2d(
+        self.convolutional_layer = nn.Conv3d(
             in_channels=parameters['input_channels'],
             out_channels=parameters['output_channels'],
             kernel_size=(parameters['kernel_depth'],
@@ -333,6 +334,7 @@ class ConvolutionalBlock3D(nn.Module):
             stride=parameters['convolution_stride'],
             padding=(padding_depth, padding_heigth, padding_width)
         )
+
         self.activation = nn.PReLU()
 
         # Other activation functions which might be interesting to test: