Defining a model class in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the PyTorch module needed to define a model.

PyTorch

import torch.nn as [1]

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

Aoptim

Butils

Cnn

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Importing 'optim' instead of 'nn' for model definition.

Using 'data' which is for datasets, not models.

The 'nn' module in PyTorch contains classes to build neural network models.

2fill in blank

medium

Complete the code to define a model class that inherits from the correct PyTorch base class.

PyTorch

class MyModel([1]):
    def __init__(self):
        super().__init__()

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

Ann.Linear

Bnn.Module

Cnn.Parameter

Dnn.Sequential

Attempts:

3 left

💡 Hint

Common Mistakes

Inheriting from nn.Sequential which is a container, not a base class.

Using nn.Linear which is a layer, not a base class.

All PyTorch models should inherit from nn.Module to work properly.

3fill in blank

hard

Fix the error in the forward method definition by completing the method signature correctly.

PyTorch

def [1](self, x):
    return self.layer(x)

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

Arun

Bpredict

Ccall

Dforward

Attempts:

3 left

💡 Hint

Common Mistakes

Naming the method 'predict' which is not recognized by PyTorch.

Using 'call' or 'run' which are not valid method names here.

The forward method must be named 'forward' for PyTorch to know how to compute outputs.

4fill in blank

hard

Fill both blanks to define a linear layer with 10 inputs and 5 outputs inside the model's __init__ method.

PyTorch

self.layer = nn.[1]([2], 5)

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

ALinear

B10

CConv2d

DReLU

Attempts:

3 left

💡 Hint

Common Mistakes

Using Conv2d which is for images, not linear layers.

Putting the output size before the input size.

The linear layer is created with nn.Linear and input size 10 for 5 output features.

5fill in blank

hard

Fill all three blanks to complete the forward method that applies the linear layer and then a ReLU activation.

PyTorch

def forward(self, x):
    x = self.layer(x)
    x = nn.[1]()([2])
    return [3]

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

ASigmoid

Bx

CReLU

Dself

Attempts:

3 left

💡 Hint

Common Mistakes

Using Sigmoid instead of ReLU as activation here.

Returning 'self' instead of the output tensor.

After the linear layer, ReLU activation is applied to x, and x is returned.