Closed yianzhongguo closed 5 years ago
Network output is always 5D
batch, channel, X, Y, Z
But it seems that the output of l_pred = Conv3DLayer(l, num_output_classes, 1, pad='same', nonlinearity=None) is 4D.
it's 5D. Every conv3D layer has 5d inputs and outputs
I know the input of conv3D layer is 5D, but what does the shape of outputs is determined by?
by the batch size of the input, num_output_classes, padding, stride and input shape
the output here is is 5D with the same batch size and 3d shape of the input and num_output_classes channels
I've made it clear, thank you very much!
@FabianIsensee Hi, sir. According to the line 130 :"l_pred = Conv3DLayer(l, num_output_classes, 1, pad='same', nonlinearity=None)" in the network_architecture.py, the output shape of the network should be 4128128*128 which is dimension 4. How can it be transformed into dimension 5 in the line 139:"l = DimshuffleLayer(l, (0, 2, 3, 4, 1))" which seems to increace dimension batch. Thank you!