Beginning Anomaly Detection Using

kernel_size: The size of the pooling window. If an integer n is given, 

then the window size of the pooling layer is 

1xn.

• 

stride: Defaults to kernel_size if nothing is passed in. If you pass in 

an integer, the pooling window moves by integer 

n amount after 

completing its pooling operation on a set of entries.

• 

padding: An integer n representing the zero padding to add on both 

sides. Default = 0.

• 

dilation: Similar to the dilation factor in the convolutional layer, 

except with max pooling. Default = 1.

• 

return_indices: If set to True, it will return the indices of the max 

values along with the outputs. Default = False.

• 

ceil_mode: If set to True, it will use ceil instead of floor to compute 

the output shape. This comes into play because of the dimensionality 

reduction involved (a kernel size of 

n will reduce dimensionality by a 

factor of 

n).

appendix B   intro to pytorch

---

381

 MaxPooling2D

torch.nn.MaxPool2d()

It applies max pooling on a 2D input. To get a better idea of how max pooling works, 

check out Chapter 

3

.

This is the list of parameters:

•