AvgPool2d in PyTorch

#python #pytorch #avgpool2d #poolinglayer

*Memos:

My post explains Pooling Layer.
My post explains AvgPool1d().
My post explains AvgPool3d().
My post explains MaxPool1d().
My post explains MaxPool2d().
My post explains MaxPool3d().
My post explains requires_grad.

AvgPool2d() can get the 3D or 4D tensor of the one or more elements computed by 2D average pooling from the 3D or 4D tensor of one or more elements as shown below:

*Memos:

The 1st argument for initialization is kernel_size(Required-Type:int or tuple or list of int). *It must be 1 <= x.
The 2nd argument for initialization is stride(Optional-Default:None-Type:int or tuple or list of int): *Memos:
- It must be 1 <= x.
- If it's None, kernel_size is set.
The 3rd argument for initialization is padding(Optional-Default:0-Type:int or tuple or list of int). *It must be 0 <= x.
The 4th argument for initialization is ceil_mode(Optional-Default:False-Type:bool).
The 5th argument for initialization is count_include_pad(Optional-Default:True-Type:bool).
The 6th argument for initialization is divisor_override(Optional-Default:None-Type:int).
The 1st argument is input(Required-Type:tensor of int or float): *Memos:
- It must be the 3D or 4D tensor of one or more elements.
- The tensor's requires_grad which is False by default is not set to True by AvgPool2d().

import torch
from torch import nn

tensor1 = torch.tensor([[[8., -3., 0., 1., 5., -2.]]])

tensor1.requires_grad
# False

avgpool2d = nn.AvgPool2d(kernel_size=1)
tensor2 = avgpool2d(input=tensor1)
tensor2
# tensor([[[8., -3., 0., 1., 5., -2.]]])

tensor2.requires_grad
# False

avgpool2d
# AvgPool2d(kernel_size=1, stride=1, padding=0)

avgpool2d.kernel_size
# 1

avgpool2d.stride
# 1

avgpool2d.padding
# 0

avgpool2d.ceil_mode
# False

avgpool2d.count_include_pad
# True

avgpool2d.divisor_override
# None

avgpool2d = nn.AvgPool2d(kernel_size=1, stride=None, padding=0, 
                         ceil_mode=False, count_include_pad=True, 
                         divisor_override=None)
avgpool2d(input=tensor1)
# tensor([[[8., -3., 0., 1., 5., -2.]]])

avgpool2d = nn.AvgPool2d(kernel_size=2, padding=1)
avgpool2d(input=tensor1)
# tensor([[[2.0000, -0.7500, 1.5000, -0.5000]]])

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1)
avgpool2d(input=tensor1)
# tensor([[[0.5556, 0.6667]]])

avgpool2d = nn.AvgPool2d(kernel_size=4, padding=2)
avgpool2d(input=tensor1)
# tensor([[[0.3125, 0.2500]]])
etc.

avgpool2d = nn.AvgPool2d(kernel_size=7, padding=3)
avgpool2d(input=tensor1)
# tensor([[[0.1224]]])
etc.

avgpool2d = nn.AvgPool2d(kernel_size=7, padding=3, divisor_override=2)
avgpool2d(input=tensor1)
# tensor([[[3.]]])

my_tensor = torch.tensor([[[8., -3., 0.],
                           [1., 5., -2.]]])
avgpool2d = nn.AvgPool2d(kernel_size=1)
avgpool2d(input=my_tensor)
# tensor([[[8., -3., 0.],
#          [1., 5., -2.]]])

avgpool2d = nn.AvgPool2d(kernel_size=2)
avgpool2d(input=my_tensor)
# tensor([[[2.7500]]])

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1)
avgpool2d(input=my_tensor)
# tensor([[[1.2222]]])
etc.

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1, divisor_override=2)
avgpool2d(input=my_tensor)
# tensor([[[5.5000]]])

my_tensor = torch.tensor([[[8.], [-3.], [0.]],
                          [[1.], [5.], [-2.]]])
avgpool2d = nn.AvgPool2d(kernel_size=1)
avgpool2d(input=my_tensor)
# tensor([[[8.], [-3.], [0.]],
#         [[1.], [5.], [-2.]]])

avgpool2d = nn.AvgPool2d(kernel_size=2, padding=1)
avgpool2d(input=my_tensor)
# tensor([[[2.0000], [-0.7500]],
#         [[0.2500], [0.7500]]])

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1)
avgpool2d(input=my_tensor)
# tensor([[[0.5556]], [[0.6667]]])
etc.

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1, divisor_override=2)
avgpool2d(input=my_tensor)
# tensor([[[2.5000]], [[3.0000]]])

my_tensor = torch.tensor([[[[8.], [-3.], [0.]],
                           [[1.], [5.], [-2.]]]])
avgpool2d = nn.AvgPool2d(kernel_size=1)
avgpool2d(input=my_tensor)
# tensor([[[[8.], [-3.], [0.]],
#          [[1.], [5.], [-2.]]]])

avgpool2d = nn.AvgPool2d(kernel_size=2, padding=1)
avgpool2d(input=my_tensor)
# tensor([[[[2.0000], [-0.7500]],
#          [[0.2500], [0.7500]]]])

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1)
avgpool2d(input=my_tensor)
# tensor([[[[0.5556]], [[0.6667]]]])
etc.

avgpool2d = nn.AvgPool2d(kernel_size=3, padding=1, divisor_override=2)
avgpool2d(input=my_tensor)
# tensor([[[[2.5000]], [[3.0000]]]])

my_tensor = torch.tensor([[[[8], [-3], [0]],
                           [[1], [5], [-2]]]])
avgpool2d = nn.AvgPool2d(kernel_size=1)
avgpool2d(input=my_tensor)
# tensor([[[[8], [-3], [0]],
#          [[1], [5], [-2]]]])

DEV Community

AvgPool2d in PyTorch

Top comments (0)

Read next

Understanding Python Decorators: A Deep Dive

Python Control Flow: If, Elif, Else, and Loops Without Going Loopy

Building a Document Retrieval & Q&A System with OpenAI and Streamlit

12 tips on how to think like a web scraping expert