Number of layers of the model - neural-network

Question

I'm training a CNN. For reporting purposes, I want to find out the number of layers my model has.
From what I see in the code below, I have a total of 6 layers, layer1, layer2, conv2_drop, fc1, fc2, fc3. Am I right?
Net(
(layer1): Sequential(
(0): Conv2d(3, 10, kernel_size=(5, 5), stride=(1, 1))
(1): BatchNorm2d(10, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU()
(3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(4): Dropout(p=0.2, inplace=False)
)
(layer2): Sequential(
(0): Conv2d(10, 20, kernel_size=(5, 5), stride=(1, 1))
(1): BatchNorm2d(20, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU()
(3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(4): Dropout(p=0.2, inplace=False)
)
(conv2_drop): Dropout2d(p=0.5, inplace=False)
(fc1): Linear(in_features=500, out_features=250, bias=True)
(fc2): Linear(in_features=250, out_features=50, bias=True)
(fc3): Linear(in_features=50, out_features=10, bias=True)
)
Does my NN have 6 layers, or layer1 and layer2 themselves add 4 more each? That would be 14 layers total?

Answer 1

It is not entirely clear what one would consider a layer (is flattening a layer or just an operation? What if it's implemented as torch.nn.Module?). Neural networks are, in essence, graphs performing operations, layers are a helpful abstraction which helps us reason about them.
In PyTorch, IMO, it would be a single instance of torch.nn.Module, so in this case it would be 14. Even more so, BatchNorm is considered a layer universally and for ease I would consider others (ReLU, MaxPool2d, Dropout) as ones as well.
layer1 and layer2 is more of a block (stack of layers) in this case. Still your fc1, fc2 and fc3 should be IMO (notice I've used it again) coded as another block (wrapped in torch.nn.Seqeuntial probably).