nn.MaxPool2d and nn.AvgPool2d in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a max pooling layer with a kernel size of 2.

PyTorch

import torch.nn as nn

pool = nn.[1](kernel_size=2)

Drag options to blanks, or click blank then click option'

AConv2d

BAvgPool2d

CMaxPool2d

DLinear

Attempts:

3 left

💡 Hint

Common Mistakes

Using nn.AvgPool2d instead of nn.MaxPool2d

Using nn.Conv2d which is for convolution, not pooling

The nn.MaxPool2d layer performs max pooling, which selects the maximum value in each window of the input.

2fill in blank

medium

Complete the code to create an average pooling layer with a stride of 2.

PyTorch

import torch.nn as nn

pool = nn.AvgPool2d(kernel_size=2, stride=[1])

Drag options to blanks, or click blank then click option'

A1

B3

C4

D2

Attempts:

3 left

💡 Hint

Common Mistakes

Setting stride to 1 causes overlapping pooling windows

Using stride larger than kernel size can skip input regions

The stride controls how far the pooling window moves. Setting stride=2 moves the window by 2 pixels each step.

3fill in blank

hard

Fix the error in the code to apply max pooling to input tensor x.

PyTorch

import torch
import torch.nn as nn

x = torch.randn(1, 1, 4, 4)
pool = nn.MaxPool2d(kernel_size=2)
output = pool([1])

Drag options to blanks, or click blank then click option'

Ax

Bpool

Cnn

Dtorch

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the pooling layer itself instead of the input tensor

Passing the torch module instead of the input tensor

The pooling layer is a function that takes the input tensor x and returns the pooled output.

4fill in blank

hard

Fill both blanks to create an average pooling layer with kernel size 3 and stride 1.

PyTorch

import torch.nn as nn

pool = nn.[1](kernel_size=[2], stride=1)

Drag options to blanks, or click blank then click option'

AAvgPool2d

BMaxPool2d

C3

D2

Attempts:

3 left

💡 Hint

Common Mistakes

Using MaxPool2d instead of AvgPool2d

Setting kernel size to 2 instead of 3

Use AvgPool2d for average pooling. The kernel size is set to 3 to cover 3x3 windows.

5fill in blank

hard

Fill all three blanks to create a max pooling layer with kernel size 2, stride 2, and padding 0.

PyTorch

import torch.nn as nn

pool = nn.[1](kernel_size=[2], stride=[3], padding=0)

Drag options to blanks, or click blank then click option'

AMaxPool2d

B2

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using AvgPool2d instead of MaxPool2d

Setting stride or kernel size incorrectly

Adding padding when not needed

The max pooling layer uses MaxPool2d with kernel size and stride both set to 2. Padding is zero to avoid adding extra pixels.