1

An Internship Report
On
Python for Data Science

Submitted in partial fulfillment of the requirement for the

Degree of Bachelor of Technology

(SESSION 2024-2025)
DEPARTMENT OF ARTIFICIAL INTELLIGENCE & DATA
SCIENCE
CHAMELI DEVI GROUP OF INSTITUTIONS, INDORE

Submitted by:
Rohan Bhati (0832AD211046)

Submitted to:
Mr. Shreyas Pagare

RAJIV GANDHI PROUDYOGIKI

VISHWAVIDYALAYA BHOPAL (M.P)
1
 2

CHAMELI DEVI GROUP OF INSTITUTIONS,

INDORE

CERTIFICATE OF ORIGINALITY OF WORK

I Rohan Bhati Enrollment No. 0832AD11046 Student of the Department of

Artificial Intelligence & Data Science Branch VII Semester, have undergone the
4 weeks internship at NPTEL Python for data science. I have worked on the one
shot image recognition using Siamese neural network project during my internship
period.

Rohan Bhati
0832AD211046
VII Semester

2
 3

CHAMELI DEVI GROUP OF INSTITUTIONS,

INDORE

CERTIFICATE
This is to certify that Mr. Rohan Bhati of B.Tech (Department of Artificial
Intelligence & Data Science) Enrollment No. 0832AD211046 has completed/part
completed his Internship of 4 weeks from date 22/01/2024 to 21/02/2024 at
NPTEL Python for Data Science during the Academic Year 2024-2025 as partial
fulfillment of the B.Tech (Artificial Intelligence & Data Science) course.

Training Coordinator Head of the Department

3
 4

Attached Certificate

Table of Content
4
 5

1.0 Introduction...................................................................................................................6

2.0 Contents of Report........................................................................................................7

2.1 Front Cover (Title Page) ..................................................................................7
2.2 Declaration.........................................................................................................8
2.3 Acknowledgements............................................................................................9
2.4 Abstract..............................................................................................................10
2.6 Weekly Records ................................................................................................11
2.7 Text (Body of the Industrial Training Report) ..............................................11

3.0 Body of Report …………………….............................................................................12

3.1 Introduction/Learning Outcome .....................................................................12
3.2 Detail of Working Experience..........................................................................13
3.2.1 Description of Technologies ………………...........................................13
3.2.2 Description of Task/Modules .................................................................14
3.2.3 Modules Snapshots ........................................................... ...................15-16
3.3 Conclusion .........................................................................................................17

1.0 Introduction
5
 6

In the era of digital transformation, data science has emerged as a critical field for unlocking
actionable insights from vast datasets, enabling organizations to make informed and strategic
decisions. To develop my capabilities in this domain, I pursued the "Python for Data Science"
certification, a rigorous program by SWAYAM NPTEL in collaboration with IIT Madras. This
certification not only equipped me with the theoretical foundations of data science and machine
learning but also provided hands-on training in implementing models that address real-world
problems.

The course covered essential data science tools, such as NumPy for efficient numerical
computations and Pandas for advanced data manipulation and analysis, enabling me to work
seamlessly with large datasets. With scikit-learn, I developed a range of machine learning
models, mastering techniques in supervised learning (e.g., regression and classification),
unsupervised learning (e.g., clustering), and semi-supervised learning, which is particularly
valuable in scenarios with limited labeled data. Each module in the certification was designed to
build core competencies, ensuring practical application of theoretical concepts in a variety of
contexts, from exploratory data analysis to predictive modeling.

One of the highlights of this certification was the capstone project, where I designed and
implemented a Siamese Neural Network (SNN) for one-shot image recognition using
TensorFlow. This project underscored the power of deep learning in complex pattern recognition
tasks. Unlike traditional models, the Siamese Neural Network is capable of identifying
similarities between images with minimal training data, making it ideal for applications such as
facial recognition, document verification, and anomaly detection. This project not only deepened
my understanding of neural network architectures but also enhanced my ability to work with
advanced machine learning frameworks and optimize models for performance.

The knowledge and skills acquired through this certification have provided a strong foundation
for my journey in data science. By mastering these tools and techniques, I am now well-prepared
to address the challenges of modern data-driven environments. This report details my learning
outcomes, the tools and methods I employed, and the real-world implications of my work,
underscoring my commitment to applying data science to drive impactful solutions.

2.0 Contents of Internship Report

6
 7

2.1 Front Cover (Title Page)

CHAMELI DEVI GROUP OF INSTITUTIONS, INDORE(M.P.)

DEPARTMENT OF ARTIFICIAL INTELLIGENCE AND

DATA SCIENCE

An Internship Report
On
Data Science
(Durarion 15 March, 2023 to 15 April 2023)

College Training Coordinator Submitted By:

Mr. Shreyas Pagare Rohan Bhati (0832AD211046)
Assistant Professor AD 5th sem/ 3rd year

2.2 Declaration

7
 8

I, Rohan Bhati, student of B.Tech VII semester in the department of Artificial Intelligence &
Data Science at Chameli Devi Group of Institutions, Indore hereby declare that the Internship
work entitled “NPTEL Python for Data Science” has been independently carried out by me
under the supervision of Mr. Deepesh Shrivas, Assistant Professor, and the coordinator Mr.
Shreyas Pagare Assistant Professor department of AD submitted in partial fulfillment of the
course requirement for the award of degree in Bachelor of Technology in Artificial Intelligence
& Data Science of RGTU University, Bhopal during the year 2024-25. I further declare that the
report has not been submitted to any other University for the award of any other degree.

Rohan Bhati
0832AD211046

2.3 Acknowledgement

I would like to express my sincere gratitude to everyone who supported and guided me during my
internship and the "Python for Data Science" certification from SWAYAM NPTEL, conducted
8
 9

by IIT Madras. This journey would not have been possible without the encouragement and
mentorship of several individuals.

First and foremost, I would like to thank Prof. Raghunathan Rengasamy for his continuous
support, guidance, and invaluable mentorship throughout my academic journey. His dedication to
fostering a strong foundation in data science has been instrumental in my development and
success in this program.

I am especially grateful to the faculty and course instructors at IIT Madras for their expert
teaching and well-structured course materials, which provided me with a comprehensive
understanding of data science and machine learning. Their insights and hands-on approach
greatly enhanced my learning experience and prepared me to tackle real-world challenges in
artificial intelligence and data analytics.

2.4 Abstract
This internship report details my experience completing the "Python for Data Science"
certification through SWAYAM NPTEL, in collaboration with IIT Madras, which provided me

9
 10

with foundational and advanced skills in data science and machine learning. The program
emphasized practical applications of Python libraries, such as NumPy for numerical
computations and Pandas for data manipulation, equipping me with the ability to handle and
process large datasets efficiently. Additionally, I explored machine learning models using
scikit-learn, covering various techniques like supervised learning, unsupervised learning, and
semi-supervised learning, enhancing my understanding of predictive modeling in diverse
contexts.

A major component of the certification was the capstone project, where I developed a Siamese
Neural Network (SNN) for one-shot image recognition, a specialized task in deep learning that
aims to identify similar images with minimal training data. Implemented using TensorFlow, the
Siamese Network project introduced me to neural network architectures tailored for high-
accuracy recognition, even with limited labeled data. This project was particularly significant as
it provided a hands-on experience in building a robust model for real-world applications, such
as facial recognition or verification systems, and demonstrated the versatility of deep learning in
specialized tasks.

Throughout the course, I faced challenges like optimizing model performance and balancing
computational resources, which were invaluable for building problem-solving skills in data
science. This report outlines the skills gained, projects completed, and the broader impact of this
certification on my professional journey in artificial intelligence and data science. The
experience has not only strengthened my technical capabilities but has also prepared me to
apply these tools in future projects that demand data-driven solutions and advanced analytics.

2.5 Weekly Records

10
 11

Date Task Skills

15/03/2023 Iris Flower Classification Analytics, Data Preprocessing,

Training, Classification, SVM
22/03/2023 Unemployment Analysis with Analytics
Python
29/03/2023 Car Price prediction with Analytics, Decision Tree
Machine Learning Regressor
05/04/2023 Email Spam detection with Analytics, Classification,
Machine Learning Random Forest Classifier, Count
Vectorizer
12/04/2023 Sales prediction using Python Analytics, Linear Regression

Students are required to record their Weekly task in a log book. Students may summarize all
the assigned tasks of each week. Student should include vital information pertaining to the
development of the necessary skill sets, e.g. programming skills, analysis skills, writing
skills, etc.

In summary, Weekly Records sheet records a Weekly tasks assigned and

experiences, observations, knowledge gained throughout the internship period.

Refer to sample in Appendix E (for engineering students).

2.7 Text (Body of the Internship Report)

An Internship Report must be divided into chapters. A title must be given to each chapter
which reflects its content. A new chapter must begin on a new page. A chapter must be further
divided into different sections with appropriate titles numbered accordingly.

The body of the Internship Report must be written in paragraphs. Each paragraph
describes an issue or a subject. There must be continuity or logical flow between paragraphs.
Long paragraphs should be avoided.

The text should contain the following:-

‹ Introduction/Learning Outcomes
‹ Details of The Working Experience
‹ Conclusions

Refer to Section 3.0 for the details on each of the topics above.

3.0 Body of Internship Report

11
 12

3.1 Learning Outcome

The "Python for Data Science" certification provided an in-depth understanding of data science
tools, machine learning techniques, and deep learning applications, contributing significantly to
my technical and analytical capabilities. Key learning outcomes from the course are outlined
below:
1. Proficiency in Data Science Libraries:
Gained expertise in NumPy and Pandas for data handling, cleaning, and analysis, which
allowed me to manipulate large datasets efficiently. These libraries enhanced my skills in data
preprocessing, including filtering, aggregation, and feature engineering.

2. Understanding of Machine Learning Models:

Acquired a comprehensive understanding of supervised learning techniques, such as regression
and classification, and unsupervised learning methods like clustering. These methods provided a
foundation for developing predictive models and extracting insights from unlabeled data.

3. Project Development in Deep Learning:

Designed and implemented a Siamese Neural Network (SNN) for one-shot image
recognition, gaining hands-on experience in deep learning model architecture and practical
applications. This project deepened my understanding of convolutional neural networks (CNNs)
and how they can be adapted for unique recognition tasks.

Learned to work with TensorFlow for building and fine-tuning neural networks, which
provided insights into handling image data, designing loss functions for similarity measurement,
and applying backpropagation to optimize network weights.

4. Model Evaluation and Optimization:

Gained skills in model evaluation metrics, such as accuracy, precision, recall, and F1-score,
learning how to interpret these metrics to assess model performance effectively.

5. Problem-Solving and Analytical Thinking:

Developed problem-solving skills by addressing challenges during model development, such as
computational constraints, data preprocessing bottlenecks, and performance tuning.
Enhanced my ability to critically analyze model outputs, debug issues, and apply iterative
testing to improve model efficiency, an essential skill for real-world data science projects.

6. Industry-Ready Skills:
The hands-on experience with popular libraries and frameworks has prepared me to handle real-
world data science projects. I am now able to work effectively with Python and industry-
standard tools, making me capable of contributing to projects requiring data analysis, machine
learning, and AI-driven solutions.

Name of organization : IIT Madras

Name and position of industrial supervisor : Prof. Raghunathan Rengasamy
Start date and completion data : 22/01/2024 to 21/02/2024

3.2 Detail of Working Experience

3.2.1 Description of Technologies

12
 13

NumPy: A fundamental Python library for numerical computations, essential for handling
large, multi-dimensional arrays and matrices. It provides mathematical functions to perform
operations on these arrays, making it a core library for scientific computing and data analysis
tasks.

Pandas: A powerful Python library for data manipulation and analysis, especially useful for
handling structured data in the form of tables. Pandas provides data structures like DataFrames,
which allow for easy data cleaning, transformation, and analysis, making it indispensable for
data science projects.

Matplotlib: A comprehensive library for creating static, animated, and interactive visualizations
in Python. It enables data scientists to create various charts, plots, and graphs to visualize and
interpret data trends, helping in better understanding and presenting data insights.

Scikit-Learn (sklearn): A machine learning library for Python that provides simple and
efficient tools for data mining and analysis. It includes algorithms for classification, regression,
clustering, dimensionality reduction, and more, making it a go-to tool for building machine
learning models quickly and effectively.

Deep Learning: A subset of machine learning that utilizes neural networks with multiple layers
to model complex patterns in large data sets. Deep learning is key for tasks like image
recognition, natural language processing, and facial recognition, as it allows models to learn
intricate data representations.

TensorFlow: An open-source deep learning framework developed by Google, widely used for
building and training neural networks. TensorFlow supports large-scale machine learning and
deep learning applications, providing tools for model building, data processing, and
deployment, and is a popular choice for both research and production in AI projects.

3.2.2 Description of Tasks/ Modules

Task 1: Data Visualization on Titanic Dataset

13
 14

Done data visualization task on the titanic dataset to understand the patterns of data like , how
many peoples survived in disaster, from which class, how many males and females survived,
which class peoples how much survived, frequency of ages of peoples to track which age
people travelled in ship and various data visualization performed.

Task 2: Data Preprocessing on Email Spam Dataset

Performed data cleaning and preprocessing on email spam dataset, encoding the variables,
rename the features, duplicated values, missing values etc. Then I used nltk library to clean
text data by stopping stop words like ‘I’, ‘me’, ‘myself’, ‘we’, ‘our’ ‘ourselves’ etc.. After this
I used string library to discard punctuation from the data like '!"#$%&\'()*+,-./:;<=>?@[\\]^_`
etc. then again using nltk PorterStemmer transform the text example dancing, dancer converted
into danc , loving, lovabe into lov.

Task 3 : Siamese Neural Network for one shot image recognition

Created a Siamese Neural Network to classify the image of person is verified or unverified,
first I taken the unverified peoples images downloaded from online, and multiple image of
single verified people then I preprocess the data in proper format to train a model, this model is
based on CNN here we take 105x105 pixel image and pass on multiple hidden layer same
process we do with verified and unverified person image after passing all the layers images we
check the input with the testing image and calculate the l1 distance that is how much both
image is similar if they are similar it means image is verified otherwise unverified.

3.2.3 Modules Snapshots

14
 15

Task 1 :

Task 2 :

Task 3 :

15
 16

3.3 Conclusion

16
 17

Completing the "Python for Data Science" certification through SWAYAM NPTEL in
collaboration with IIT Madras has been a transformative experience, providing me with both
theoretical and practical expertise in data science and machine learning. The course not only
deepened my understanding of foundational data science tools like NumPy and Pandas but
also enabled me to apply these tools effectively in data preprocessing and analysis.
Furthermore, the certification covered a range of machine learning algorithms through scikit-
learn, helping me master predictive modeling and clustering techniques that are essential in
real-world applications.

One of the most impactful aspects of this certification was the capstone project, where I
developed a Siamese Neural Network for one-shot image recognition using TensorFlow.
This project gave me valuable insights into deep learning, specifically the architecture and
applications of neural networks for tasks that require high precision with minimal data.
Overcoming challenges in model design and optimization during this project sharpened my
problem-solving skills and taught me the importance of model evaluation and iterative
improvement.

The knowledge and skills gained through this certification have prepared me to take on
complex challenges in the field of artificial intelligence and data science. I am now equipped
to contribute effectively to projects involving data analytics, machine learning, and advanced
AI, bringing both technical expertise and analytical rigor. This internship experience has laid
a solid foundation for my career in data science, and I am eager to apply these skills in future
projects, driving innovation and creating impactful data-driven solutions.
.

17