You are on page 1of 6

GOVERNMENT COLLEGE OF ENGINEERING, JALGAON

(An Autonomous Institute of Govt. of Maharashtra)


Department of Computer Engineering

CERTIFICATE

This is to certify that the Seminar report entitled, “Keras”, which is being submitted
herewith for the award of L.Y. B.TECH., is the result of the work completed under my
supervision and guidance within the four walls of the institute and the same has not been
submitted elsewhere for the award of any degree.

Prof. H.D. Gadade (Prof.D.V.Chaudhari)


(Guide) Head of Department

Principal Examiner

i
DECLARATION

I hereby declare that the seminar report entitled, “Keras” was carried out and written
by me under the guidance of Prof. H.D. Gadade, Department of Computer Engineering,
Government College of Engineering, Jalgaon. This work has not been previously formed the
basis for the award of any degree or diploma or certificate nor has been submitted elsewhere
for the award of any degree.

Place: Jalgaon
URVASHI S RANE (1641047)
Date: / /2019

ACKNOWLEDGEMENT

ii
A successful work makes an important event in the course of college study. I have
great pleasure submitting seminar report on “Keras”.

Firstly, I wish to express my deep scene of gratitude to my guide Prof. H.D Gadade ,
for immense interest, valuable guidance, constant inspiration and kind co-operation through
the period of work undertaken which has been very instrumental in process of my work.

I am very much thankful to our Head of Department Prof. D.V.Chaudhari, who has
always encouraged and inspired us, because of whom we could learn to, manage the things in
required time.

I would thank to our Principal Dr. R.D.Kokate, for his kind support and guidance,
they set a path for performing and inventing this project report.

URVASHI S RANE(1641047)

INDEX

iii
Abstract 1

1 Introduction 2
1.1 Keras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Deep learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.3 Artificial Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.4 Multi-Layer Perceptrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.5 Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.6 Organization Of report. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2 Literature Survey 7

2.1 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.3 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.4 Disadvantages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3 Design 11
3.1 Why use Keras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.1 Keras has broad adoption in the industry and the research community 11

3.1.2 Keras makes it easy to turn models into products . . . . . . . . . . . 12


3.1.3 Keras supports multiple backend engines and does not lock you into
one ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3.1.4 Keras has strong multi-GPU support and distributed training support 12
3.1.5 Keras development is backed by key companies in the deep learning
ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2 Multi-backend Keras and tf.keras: . . . . . . . . . . . . . . . . . . . . . . . . 13

3.3 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.4 Disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

4 Implementation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 15

4.1 Building CNN using Keras. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15


5 Conclusion 22

iv
References 23

List of Figures

v
1.1 Visual Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

3.1 Deep learning frameworks ranking computed by Je Hale, based on 11 data 11


sources across 7 categories . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Setting up the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2 Installing Keras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.3 Importing libraries and modules . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.4 Load image data from MNIST . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.5 Preprocess input data for Keras . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.6 Preprocessing class labels for Keras . . . . . . . . . . . . . . . . . . . . . . . 18
4.7 Printing the shape of current model output . . . . . . . . . . . . . . . . . . . 19
4.8 De ning model architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.9 Compile model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.10 Fitting model on training data . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.11 Evaluating model on test data . . . . . . . . . . . . . . . . . . . . . . . . . . 21

vi

You might also like