__init__ for layers in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define the constructor of a PyTorch layer class.

PyTorch

class MyLayer(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear([1], 10)

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

ANone

B10

C0

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Using None or 0 as input size causes errors.

Confusing input and output sizes.

The input size to nn.Linear must be specified. Here, 5 is the input feature size.

2fill in blank

medium

Complete the code to initialize a convolutional layer with 3 input channels and 16 output channels.

PyTorch

class ConvLayer(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d([1], 16, kernel_size=3)

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

A16

B32

C3

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using output channels as input channels.

Using kernel size as input channels.

The first argument to nn.Conv2d is the number of input channels, here 3 for RGB images.

3fill in blank

hard

Fix the error in the layer initialization by filling the blank correctly.

PyTorch

class CustomLayer(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.fc = nn.Linear(input_dim, [1])

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

Aoutput_dim

Binput_dim

Cinput_dim * 2

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Using input_dim for both input and output sizes.

Using None as output size.

The second argument to nn.Linear is the output dimension, which should be output_dim, not input_dim.

4fill in blank

hard

Fill both blanks to initialize a dropout layer with 0.5 dropout rate and a ReLU activation.

PyTorch

class DropoutReluLayer(nn.Module):
    def __init__(self):
        super().__init__()
        self.dropout = nn.Dropout([1])
        self.activation = nn.[2]()

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

A0.5

B0.3

CReLU

DSigmoid

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong dropout rate like 0.3.

Using Sigmoid instead of ReLU.

Dropout rate is 0.5 and activation is ReLU as requested.

5fill in blank

hard

Fill all three blanks to define a layer with batch normalization, dropout, and linear transformation.

PyTorch

class ComplexLayer(nn.Module):
    def __init__(self, input_size, output_size):
        super().__init__()
        self.batchnorm = nn.BatchNorm1d([1])
        self.dropout = nn.Dropout([2])
        self.linear = nn.Linear([3], output_size)

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

Ainput_size

B0.2

Doutput_size

Attempts:

3 left

💡 Hint

Common Mistakes

Using output_size for batchnorm or linear input.

Using dropout rate greater than 1.

BatchNorm1d uses input_size, dropout rate is 0.2, and linear layer input is input_size.