nn.Conv2d layers in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a 2D convolutional layer with 3 input channels and 16 output channels.

PyTorch

conv_layer = nn.Conv2d([1], 16, kernel_size=3)

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

A1

B3

C16

D32

Attempts:

3 left

💡 Hint

Common Mistakes

Using the number of output channels as the first argument.

Confusing kernel size with input channels.

The first argument to nn.Conv2d is the number of input channels. Here, it should be 3.

2fill in blank

medium

Complete the code to apply the convolutional layer to the input tensor.

PyTorch

output = conv_layer([1])

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

Ainput_tensor

Bconv_layer

Coutput

Dnn.Conv2d

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the layer itself instead of the input tensor.

Using the output variable before it is defined.

You apply the convolutional layer to the input tensor by calling it like a function with the input tensor as argument.

3fill in blank

hard

Fix the error in the code by completing the missing argument for padding.

PyTorch

conv_layer = nn.Conv2d(3, 16, kernel_size=3, padding=[1])

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

A1

B0

C2

D3

Attempts:

3 left

💡 Hint

Common Mistakes

Using padding 0 causes output size to shrink.

Using padding larger than needed causes unexpected output size.

Padding of 1 keeps the output size the same when using a 3x3 kernel.

4fill in blank

hard

Fill both blanks to create a convolutional layer with stride 2 and kernel size 5.

PyTorch

conv_layer = nn.Conv2d(3, 32, kernel_size=[1], stride=[2])

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

A5

B2

C3

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Confusing stride with padding.

Using kernel size 3 instead of 5.

The kernel size is 5 and stride is 2 to downsample the input.

5fill in blank

hard

Fill all three blanks to create a Conv2d layer with 1 input channel, 10 output channels, kernel size 3, padding 1.

PyTorch

conv_layer = nn.Conv2d([1], [2], kernel_size=[3], padding=1)

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

A1

B10

C3

D16

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up input and output channels.

Forgetting to set padding to 1 for same output size.

The layer has 1 input channel, 10 output channels, and kernel size 3 with padding 1.