DEEP LEARNING USING KERAS AND TENSORFLOW

20IT007 2/0/2/3
(COMMON TO CSE / IT)
Nature of Course M (Practical Application)
Pre requisites Python Programming, Machine Learning Techniques
Course Objectives:
To become familiar with the language and fundamental concepts of artificial
1.
neural networks.
2. To understand and implement Deep Learning Architectures.
To familiarize the student with the Image Processing facilities like TensorFlow
3.
and Keras.
To get familiar with the use of TensorFlow/Keras in Deep Learning
4.
Applications.
To make the students build deep learning models, interpret results, and build
5.
own deep learning projects.
Course Outcomes
Upon completion of the course, students shall have ability to
Understand and review tools available to build Deep Learning
C007.1 [U]
including Tensor Flow, Keras.
Build Deep Learning Machine Learning models using
C007.2 [AP]
TensorFlow and various interfaces
Implement deep neural network models to generate realistic
C007.3 [AP]
images in Tensorflow
Develop probabilistic models with TensorFlow, making particular
C007.4 [AP]
use of the TensorFlow Probability library
Classify regression and classification models using the Keras
C007.5 library together with convolutional networks and to build them [A]
using the Keras library
Discover and develop deep learning for computer vision
C007.6 [AP]
problems using Keras
Course Contents:

Introduction to Artificial Neural Networks with Keras: 15

Hours
From Biological to Artificial Neurons - Implementing MLPs with Keras: Building an
Image Classifier Using the Sequential API - Building an Image Classifier and regression
MLP using the Sequential API - Building Complex Models using the Functional API -
Building Dynamic Models Using the Subclassing API – Fine-Tuning Neural Network
Hyperparameters. Training Deep Neural Networks: Vanishing/Exploding Gradients
Problems - Reusing Pretrained Layers - Faster Optimizers - Avoiding Overfitting through
Regularization.

Models and Training with Tensorflow: 15

Hours
Tour of TensorFlow - Using TensorFlow like NumPy -Customizing Models and Training
Algorithms - TensorFlow Functions and Graphs - Loading and Preprocessing Data
with TensorFlow: The Data API - The TFRecord Format - The Features API - TF
Transform. Case Study: The TensorFlow Datasets (TFDS) Project

Deep Computer Vision using Convolutional Neural Networks: 15

Hours
The Architecture of the Visual Cortex - Convolutional Layer - Pooling Layer - CNN
Architectures: LeNet-5 – AlexNet – GoogLeNet – VGGNet – ResNet – Xception – SENet.
Implementing a ResNet-34 CNN using Keras - Using Pretrained Models from Keras -
Pretrained Models for Transfer Learning - Classification and Localization - Object
Detection - Semantic Segmentation
Total Hours 45 hours
Laboratory Component:
S. No List of Experiments
1. Heart Disease Classification using Keras
2. Face Mask Detection using Keras
3. Detecting Spam using TensorFlow.

4. Image Classification with TensorFlow

5. Optical Character Recognition using TensorFlow

Total Hours 30 hours

Text Books:
Aurélien Géron, “Hands-On Machine Learning with Scikit-Learn, Keras, and
1. TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems”, 2nd
Edition, O'Reilly Media, 2019.
Antonio Gulli, Amita Kapoor, Sujit Pal “Deep Learning with TensorFlow 2 and
2.
Keras”, 2nd Edition, Packt Publishing, 2019
Ian Goodfellow, Yoshua Bengio, Aaron Cour, “Deep Learning (Adaptive
3.
Computation and Machine Learning series)”, The MIT Press , 2016.
Reference Books:
Liangqu Long, Xiangming Zeng, “Beginning Deep Learning with TensorFlow: Work
1. with Keras, MNIST Data Sets, and Advanced Neural Networks”, APress; 1st Edition,
2022
Joseph, Ferdin Joe John, Sarayut Nonsiri, and Annop Monsakul. "Correction to
2. Keras and TensorFlow: A Hands-On Experience." Advanced Deep Learning for
Engineers and Scientists. Springer, 2021.
Davies, E. R., and Matthew Turk. “Advanced methods and deep learning in
3.
computer vision”, Elsevier Science & Technology, 2021.
Web References:
1. https://www.kaggle.com/jameskhoo/deep-learning-with-keras-and-tensorflow
2. https://codebasics.io/courses/deep-learning-with-tensorflow-keras-and-python
3. https://github.com/codebasics/deep-learning-keras-tf-tutorial

Online Resources:
1. https://www.edx.org/course/deep-learning-with-tensorflow
2. https://datascience.uci.edu/education/data-science-short-courses/
3. https://onlinecourses.nptel.ac.in/noc19_cs81/preview

Tentative Assessment Methods & Levels (based on Revised Bloom’s Taxonomy)

Summative assessment based on Continuous and End Semester Examination
Continuous Assessment
End Semester
Revised Theory Practical
Examination
Bloom’s Rubric
CIA-1 CIA-2 CIA-3 (Theory)
Level based CIA
[10 marks] [10 marks] [10 marks] [40 marks]
[30 Marks]
Remember - - - - -
 Understand 30 30 30 20 30
Apply 40 60 70 60 60
Analyse 30 10 - 20 10
Evaluate - - - - -
Create - - - - -

Programme Specific
Course Outcomes Programme Outcomes (PO)
Outcomes (PSO)
(CO)
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3
C007.1 2 3 3 3 3 3 2 2
C007.2 1 2 2 2 2 3 2 1
C007.3 3 3 2 2 3 3 3 3
C007.4 1 2 3 3 1 2 2 2
C007.5 2 3 2 2 2 2 3 2
C007.6 2 2 3 2 2 3 2 3