Scribd
Upload
25 views2 pages

Class Notes DL Unit 1

The document outlines the structure and content of a mid-term examination for a Deep Learning course at MVR College of Engineering and Technology. It includes questions on various topics such as activation functions, classification types, neural network architectures, and frameworks like Keras and PyTorch. The exam is divided into two sets with questions carrying different marks and focuses on both theoretical concepts and practical coding implementations.

Uploaded by

madhu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
25 views2 pages

Class Notes DL Unit 1

The document outlines the structure and content of a mid-term examination for a Deep Learning course at MVR College of Engineering and Technology. It includes questions on various topics such as activation functions, classification types, neural network architectures, and frameworks like Keras and PyTorch. The exam is divided into two sets with questions carrying different marks and focuses on both theoretical concepts and practical coding implementations.

Uploaded by

madhu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

MVR COLLEGE OF ENGINEERING AND TECHNOLOGY:PAITALA

AUTONOMOUS

III_ B.Tech _II_ Sem (RM21) Course Code:21CS32304


Date: 23 /04/2025
AY: 2024-2025 Course Title: DEEP LEARNING

Programme : CSE-AI A ,AI-B,AIDS MID-II SET-I Duration:2hrs Max Marks:25

Note: Answer all of the following Max BTL


(Each Question Carries 2 Marks) 5*2=10 Marks CO (1 To 6)

Q1 Purpose of the softmax() activation function in Keras? 1


A 2M 2
What is the difference between binary classification and 1
B 2M 1
multiclass classification?
What is the difference between a CNN and an RNN? 2
C 2M 2

primary function of Generative Adversarial Networks 2


D 2M 1

Application of Deep Generative Models? 2


E 2M 2

Note: Answer all of the following Max BTL


(Each Question Carries 5 Marks) 3*5=15. Marks CO (1 To 6)

Explain the concept of Generative Adversarial Networks (GANs) in 1


Q.2. deep learning. Describe the architecture of a GAN and its applications
in real-world scenarios, including image generation and data 5M 2
augmentation.

Write a PyTorch code snippet to implement a simple RNN. Explain the 1


Q.3 role of each line of code in the implementation. 5M
1

2
Q.4 Describe the process of compiling a neural network model in Keras.
Explain the role of loss function, optimizer, and evaluation metrics. 5M 2
MVR COLLEGE OF ENGINEERING AND TECHNOLOGY:PAITALA
AUTONOMOUS

III_ B.Tech _II_ Sem (RM21) Course Code: 21CS32304


Date: 23 /04/2025
AY: 2024-2025 Course Title: DEEP LEARNING

Programme : CSE-AI A ,AI-B,AIDS MID-II SET-II Duration:2hrs Max Marks:25

Note: Answer all of the following Max BTL


(Each Question Carries 2 Marks) 5*2=10 Marks CO (1 To 6)

Q1 Purpose of the compile() method in Keras? 1


A 2M 2

B What is nn.RNN() module in PyTorch? 2M 1 1

What is torch.nn.functional module in PyTorch? 2


C 2M 2

primary application of Machine Vision in deep learning? 1


D 2M 2
What is the main application of Deep Reinforcement 2
E 2M 1
Learning?

Note: Answer all of the following Max BTL


(Each Question Carries 5 Marks) 3*5=15. Marks CO (1 To 6)

Compare and contrast Keras, TensorFlow, Theano, and CNTK. 2


Q.2.
5M 1
Explain the advantages and disadvantages of each framework
Describe the architecture of a Convolutional Neural Network (CNN). 1
Q.3 Explain the role of convolutional layers, pooling layers, and fully 2
connected layers in a CNN. 5M

Introduce the concept of Deep Generative Models in deep learning.


Explain the different types of Deep Generative Models, including 2
Q.4
Variational Autoencoders (VAEs) and Generative Adversarial 5M 1
Networks (GANs).

You might also like